# Documentation Map Learn how to get started with the AI/ML API This page helps you quickly find the right AI model for your task.\ Open the API reference and copy a working example to integrate it into your code in minutes. *** **Trending Models**
Cover image
Pro-Grade Image Model/files/mhKXRNgl7WqEiJigXJNC/pages/08352f10db7ff676e27d5f6d6cacae74cacc97c4
Frontier AI Reasoning Model/files/KmRbGJYsVya9jArMMisa/pages/72c7692382912055c1b330d34fd07d4428f6bf44
Fast Video Generation/files/fAcuv79XecmCpgrwxs1m/pages/IjlHF9CtCyN0QPMtHBl3
***

Start with this code block

🚀 Setup guide

🧩 SDKs

▶️ Run in Playground

from openai import OpenAI
client = OpenAI(
base_url="https://api.aimlapi.com/v1",
api_key="<YOUR_AIMLAPI_KEY>",
)
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Write a one-sentence story about numbers."}]
)
print(response.choices[0].message.content)
*** Select the model by its **Task**, by its **Developer** or by the supported **Capabilities**: {% hint style="info" %} If you've already made your choice and know the model ID, use the [Search panel](https://docs.aimlapi.com/?q=) on your right. {% endhint %} {% tabs %} {% tab title="Models by TASK" %} {% content-ref url="/pages/6a2JwgGnh8dvD5mmdzoP" %} [Text Models (LLM)](/api-references/text-models-llm) {% endcontent-ref %} {% content-ref url="/pages/RFizWjOk4ojVgsvmejjK" %} [Image Models](/api-references/image-models) {% endcontent-ref %} {% content-ref url="/pages/LUpdTqMHZ9YRxWhhrPCV" %} [Video Models](/api-references/video-models) {% endcontent-ref %} {% content-ref url="/pages/W4MpGk1WlcYX0fNrH2IP" %} [Music Models](/api-references/music-models) {% endcontent-ref %} {% content-ref url="/pages/Xw5ohM8tnr5umNsfCWF6" %} [Voice/Speech Models](/api-references/speech-models) {% endcontent-ref %} {% content-ref url="/pages/0Yx821BRYTKgP4Xzk6hC" %} [3D-Generating Models](/api-references/3d-generating-models) {% endcontent-ref %} {% content-ref url="/pages/awkdnLHWRSqjdFPQahur" %} [Vision Models](/api-references/vision-models) {% endcontent-ref %} {% content-ref url="/pages/9xxhbHaEisZemoWFlpdd" %} [Embedding Models](/api-references/embedding-models) {% endcontent-ref %} {% endtab %} {% tab title="Models by DEVELOPER" %} **Alibaba Cloud**: [Text/Chat](/api-references/text-models-llm/alibaba-cloud) [Image](/api-references/video-models/alibaba-cloud) [Video](/api-references/image-models/alibaba-cloud) [Text-to-Speech](/api-references/speech-models/text-to-speech/alibaba-cloud) [Embedding](/api-references/embedding-models/alibaba-cloud) **Anthracite**: [Text/Chat](/api-references/text-models-llm/anthracite) **Anthropic**: [Text/Chat](/api-references/text-models-llm/anthropic) [Embedding](/api-references/embedding-models/anthropic) **Assembly AI:** [Speech-To-Text](/api-references/speech-models/speech-to-text/assembly-ai) **Baidu**: [Text/Chat](/api-references/text-models-llm/baidu) **ByteDance**: [Text/Chat](/api-references/text-models-llm/bytedance) [Image](/api-references/video-models/bytedance) [Video](/api-references/image-models/bytedance) **Cohere**: [Text/Chat](/api-references/text-models-llm/cohere) **DeepSeek**: [Text/Chat](/api-references/text-models-llm/deepseek) **Deepgram**: [Speech-To-Text](/api-references/speech-models/speech-to-text/deepgram) [Text-to-Speech](/api-references/speech-models/text-to-speech/deepgram) **ElevenLabs****:** [Text-to-Speech](/api-references/speech-models/text-to-speech/elevenlabs) [Voice Chat](/api-references/speech-models/voice-chat/elevenlabs) [Music](/api-references/music-models/elevenlabs) **Flux**: [Image](/api-references/image-models/flux) **Google**: [Text/Chat](/api-references/text-models-llm/google) [Image](/api-references/image-models/google) [Video](/api-references/video-models/google) [Music](/api-references/vision-models/ocr-optical-character-recognition/google) [Vision(OCR)](/api-references/music-models/google) [Embedding](/api-references/embedding-models/google) **Gryphe**: [Text/Chat](/api-references/text-models-llm/gryphe) **Hume AI**: [Text-to-Speech](/api-references/speech-models/text-to-speech/hume-ai) **Inworld**: [Text-to-Speech](/api-references/speech-models/text-to-speech/inworld) **Kling AI**: [Image](/api-references/image-models/kling-ai) [Video](/api-references/video-models/kling-ai) **Krea**: [Video](/api-references/video-models/krea) **LTXV**: [Video](/api-references/video-models/ltxv) **Meta**: [Text/Chat](/api-references/text-models-llm/meta) **Microsoft**: [Text-to-Speech](/api-references/speech-models/text-to-speech/microsoft) **MiniMax**: [Text/Chat](/api-references/text-models-llm/minimax) [Video](/api-references/video-models/minimax) [Music](/api-references/music-models/minimax) [Voice-Chat](/api-references/speech-models/voice-chat) **Mistral AI**: [Text/Chat](/api-references/text-models-llm/mistral-ai) [Vision(OCR)](/api-references/vision-models/ocr-optical-character-recognition/mistral-ai) **Moonshot**: [Text/Chat](/api-references/text-models-llm/moonshot) **NousResearch**: [Text/Chat](/api-references/text-models-llm/nousresearch) **NVIDIA**: [Text/Chat](/api-references/text-models-llm/nvidia) **OpenAI**: [Text/Chat](/api-references/text-models-llm/openai) [Image](/api-references/image-models/openai) [Speech-To-Text](/api-references/speech-models/speech-to-text/openai) [Embedding](/api-references/embedding-models/openai) **Perplexity**: [Text/Chat](/api-references/text-models-llm/perplexity) **PixVerse:** [Video](/api-references/video-models/pixverse) **RecraftAI**: [Image](/api-references/image-models/recraftai) **Reve**: [Image](/api-references/image-models/reve) **Runway**: [Video](/api-references/video-models/runway) **Sber AI**: [Video](/api-references/video-models/sber-ai) **Stability AI**: [Image](/api-references/image-models/stability-ai) [Music](/api-references/music-models/stability-ai) [3D-Generation](/api-references/3d-generating-models/stability-ai) **StepFun**: [Text/Chat](/api-references/text-models-llm/stepfun) **Tencent**: [Image](/api-references/image-models/tencent) [Video](/api-references/video-models/tencent) [3D](/api-references/3d-generating-models/tencent) **VEED**: [Video](/api-references/video-models/veed) **xAI**: [Text/Chat](/api-references/text-models-llm/xai) [Image](/api-references/image-models/xai) [Video](/api-references/video-models/xai) **Zhipu**: [Text/Chat](/api-references/text-models-llm/zhipu) {% endtab %} {% tab title="Text Models by CAPABILITY" %} {% content-ref url="/pages/OK6oBdkSyX2hIUJw0O0H" %} [Completion and Chat Completion](/capabilities/completion-or-chat-models) {% endcontent-ref %} {% content-ref url="/pages/wsJkOZ0nHCLLwDoDNQ8D" %} [Streaming Mode](/capabilities/streaming-mode) {% endcontent-ref %} {% content-ref url="/pages/lRPQC4QmxtpqCTIblyvz" %} [Code Generation](/capabilities/code-generation) {% endcontent-ref %} {% content-ref url="/pages/6yXmWtTfeUCPft5341aw" %} [Thinking / Reasoning](/capabilities/thinking-reasoning) {% endcontent-ref %} {% content-ref url="/pages/xhkO48FYd9TFBwAHGMmt" %} [Function Calling](/capabilities/function-calling) {% endcontent-ref %} {% content-ref url="/pages/JvoFINbW7KOduZAFvOwX" %} [Vision in Text Models](/capabilities/image-to-text-vision) {% endcontent-ref %} {% content-ref url="/pages/Ofr7iaCJpt04wnPRKfn1" %} [Web Search](/capabilities/web-search) {% endcontent-ref %} {% endtab %} {% endtabs %} *** ## Going Deeper

Use more text model capabilities in your project:

📖 ​Completion and Chat Completion

📖 Function Calling

📖 Streaming Mode

📖 Vision in Text Models (Image-to-Text)

📖 Code Generation

📖 Thinking / Reasoning

📖 Web Search

Miscellaneous:

🔗 Integrations

📗 Glossary

⚠️ Errors and Messages

❓ FAQ ​


Learn more about developer-specific features:

📖 Features of Anthropic Models
# Quickstart Access leading AI models (GPT-4o, Gemini, and others) through a single unified API. Initial setup takes just a few minutes. {% hint style="success" %} If you are a manager and simply want to test a model to evaluate its performance, for instance in content generation, the quickest approach is **to use** [**our Playground**](https://aimlapi.com/app/).\ It offers an intuitive, user-friendly interface—no coding required. Programmatic API calls are best suited for developers who want to integrate a model into their own apps. {% endhint %} *** Here, you'll learn how to start using our API in your code.\ The following steps must be completed regardless of which of our models you plan to call: * [generating an AIML API Key](#generating-an-aiml-api-key), * [choosing and preparing your development environment](#choosing-and-preparing-the-development-environment), * [making an API call](#making-an-api-call). Let's walk through an example of connecting to [the Gemma 3](/api-references/text-models-llm/google/gemma-3) model via REST API. ## G**enerating an AIML API Key**
What is an API Key? You can find your AIML API keys on the [account page](https://aimlapi.com/app/keys). An AIML API key is a credential that grants you access to our API from your code.\ It is a sensitive string that is shown **only at creation time** and should be kept confidential.\ Do not share this key with anyone, as it could be misused without your knowledge.\ If you lose it, generate a new key from your dashboard. ⚠️ Note that API keys from third-party organizations cannot be used with our API: \ you need AIML API keys. *** To start using our models in the Playground or through the API, you need to create at least one API key. You may also create separate API keys for individual employees, AI agents, teams, or specific workflows and tasks. This approach can simplify usage tracking, access management, and cost monitoring across different environments or use cases. All API keys created within your account share the same account balance.
To use the AIML API, you need an account and at least one AIML API key.\ If you have not created them yet, follow these steps: 1. [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account. 2. [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI.
Detailed instructions for managing your API keys in the dashboard can be found in the corresponding [Help Center article](https://help.aimlapi.com/article/19-how-to-create-an-api-key). *** ## Choosing the Development Environment Each language has recommended environments for running code samples.
cURL
  • REQBIN is a web-based REST client that lets you quickly run cURL requests directly in your browser, without installing any tools.
  • Git Bash (Windows) or the built-in Terminal (macOS/Linux) allow you to run cURL examples and other command-line tools locally.
Python
  • Jupyter Notebook is a popular online environment for running Python code and is the fastest option if you do not want to install anything locally.
  • Visual Studio Code (VS Code) is a lightweight and widely used code editor that supports both Python and Node.js. It is suitable for running and debugging local examples and for working on real projects.
JavaScript
{% hint style="info" %} In the examples below for cURL, JavaScript and Python, we use the [**REST API**](/quickstart/supported-sdks#rest-api). This approach works with all of our APIs, but it is not the only way to integrate. You can use [**other supported SDKs**](/quickstart/supported-sdks). {% endhint %} ## Making an API Call Based on your environment, you will call our API differently. Below are three common ways to call our API using two popular languages: **cURL** (a command-line format for making HTTP requests rather than a programming language), **Python**, and **JavaScript** (NodeJS). If you want to get started really quickly, choose one of the four expandable sections below.\ Each one contains instructions for calling our model using different tools and environments.\ The first two options are especially simple and suitable even for beginners. For completeness, the same example is explained in detail in the [Code Step-by-Step](#code-step-by-step-explanation) section. {% tabs %} {% tab title="cURL" %} ```bash curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "google/gemma-3-4b-it", "messages": [ { "role": "user", "content": "Tell me about San Francisco" } ], "temperature": 0.7, "max_tokens": 512 }' ``` {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript userPrompt = 'Tell me about San Francisco' // insert your request here async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'google/gemma-3-4b-it', messages:[ { role:'user', content: userPrompt } ], temperature: 0.7, max_tokens: 512, }), }); const data = await response.json(); const answer = data.choices[0].message.content; console.log('User:', userPrompt); console.log('AI:', answer); } main(); ``` {% endcode %} {% endtab %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests user_prompt = "Tell me about San Francisco" # insert your request here response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer " }, json={ "model":"google/gemma-3-4b-it", "messages":[ { "role":"user", "content": user_prompt } ], "temperature": 0.7, "max_tokens": 512, } ) data = response.json() answer = data["choices"][0]["message"]["content"] print("User:", user_prompt) print("AI:", answer) ``` {% endcode %} {% endtab %} {% endtabs %}
⭐ How to run a cURL example in a web-based REST client (REQBIN) Calling the API via cURL through a web service like this is the simplest and fastest method, requiring no additional libraries. However, there is a downside: cURL is not a programming language, which means it has very limited capabilities for adding logic—only API calls, no loops or conditions. You can’t even extract just the specific field with the model’s text response—cURL returns the model’s full output, as you’ll see below. *** 1\. Copy the cURL example above and paste it into a text editor, such as Notepad or Notepad++. 2\. Replace the placeholder `` with your actual AIMLAPI Key. 3\. If needed, modify the prompt (the `content` field). 4\. Copy the modified example, go to the [REQBIN](https://reqbin.com/curl) website, paste it into the designated field and click **Run**:
5\. After the model processes your request, the model’s full output will be shown directly below the input field. {% hint style="info" %} Pro tip: try experimenting with the three different ways of displaying the model’s output.\ Some are more readable than others. {% endhint %}
⭐ How to run a Python example in an online Jupyter Notebook The second fastest option, and a much more convenient choice, while offering more flexibility for customizing how the output is displayed in code. *** **1**. When you open [Jupyter Notebook](https://jupyter.org/try-jupyter/notebooks/?path=Untitled.ipynb) for the first time, select **“Python 3.13 (XPython)”** in the pop-up window to indicate the programming language kernel you will be working with:
{% hint style="info" %} In some browsers, the kernel selection may look different: {% endhint %}
2\. Enter the following command in the first cell to install the `requests` library: ```bash %pip install requests ``` Click the **Run** button in the toolbar above the cell to execute it:
3\. Paste our example into the second cell, replace the placeholder with your AIMLAPI Key, then click the **Run** button in the toolbar:
4\. After the model processes your request, the result will be shown directly below the cell:
How to run a Python example locally from the command line (without an IDE) Let's start from very beginning. We assume you already installed Python (with `venv`), if not, here a [guide for the beginners](/faq/can-i-use-api-in-python). Create a new folder for test project, name it as `aimlapi-welcome` and change to it. ```bash mkdir ./aimlapi-welcome cd ./aimlapi-welcome ``` (Optional) If you use IDE then we recommend to open created folder as workspace. On example, in Visual Studio Code you can do it with: ``` code . ``` Run a terminal inside created folder and create virtual envorinment with a command: ```shell python3 -m venv ./.venv ``` Activate created virtual environment: ```shell # Linux / Mac source ./.venv/bin/activate # Windows ./.venv/bin/Activate.bat ``` Install requirement dependencies. In our case (REST API SDK) we need only `request` library: ```shell pip install requests ``` Create new file and name it as `travel.py`: ```shell touch travel.py ``` Paste following content inside this `travel.py` and replace `` with your API key you got on [first step](#generating-an-api-key): ```python import requests user_prompt = "Tell me about San Francisco" response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"google/gemma-3-4b-it", "messages":[ { "role":"user", "content": user_prompt } ], "temperature": 0.7, "max_tokens": 512, } ) data = response.json() answer = data["choices"][0]["message"]["content"] print("User:", user_prompt) print("AI:", answer) ``` Run the application: ```shell python3 ./travel.py ``` If you done all correct, you will see following output: {% code overflow="wrap" %} ```json5 User: Tell me about San Francisco AI: San Francisco, located in northern California, USA, is a vibrant and culturally rich city known for its iconic landmarks, beautiful vistas, and diverse neighborhoods. It's a popular tourist destination famous for its iconic Golden Gate Bridge, which spans the entrance to the San Francisco Bay, and the iconic Alcatraz Island, home to the infamous federal prison. The city's famous hills offer stunning views of the bay and the cityscape. Lombard Street, the "crookedest street in the world," is a must-see attraction, with its zigzagging pavement and colorful gardens. Ferry Building Marketplace is a great place to explore local food and artisanal products, and the Pier 39 area is home to sea lions, shops, and restaurants. San Francisco's diverse neighborhoods each have their unique character. The historic Chinatown is the oldest in North America, while the colorful streets of the Mission District are known for their murals and Latin American culture. The Castro District is famous for its LGBTQ+ community and vibrant nightlife. ``` {% endcode %}
How to run a JavaScript example locally from the command line (without an IDE) We assume you already have Node.js installed. If not, here is a [guide for beginners](/faq/can-i-use-api-in-nodejs). Create a new folder for the example project: ```bash mkdir ./aimlapi-welcome cd ./aimlapi-welcome ``` Create a project file: ```bash npm init -y ``` Create a file with the source code: ```bash touch ./index.js ``` And paste the following content to the file and save it: ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'google/gemma-3-4b-it', messages:[ { role:'user', content: 'Tell me about San Francisco' // Insert your prompt here } ], temperature: 0.7, max_tokens: 256, }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` Run the file: ```bash ./index.js ``` You will see a response that looks like this: {% code overflow="wrap" %} ```json5 User: Tell me about San Francisco AI: San Francisco, located in the northern part of California, USA, is a vibrant and culturally rich city known for its iconic landmarks, beautiful scenery, and diverse neighborhoods. The city is famous for its iconic Golden Gate Bridge, an engineering marvel and one of the most recognized structures in the world. Spanning the Golden Gate Strait, this red-orange suspension bridge connects San Francisco to Marin County and offers breathtaking views of the San Francisco Bay and the Pacific Ocean. ``` {% endcode %}
*** ## Code Step-by-Step Below is a step-by-step explanation of the same API call in three variants: cURL, JavaScript, and Python. All three examples send an identical request to the `google/gemma-3-4b-it` chat model.
cURL *** **1. Command start** ```bash curl -L \ ``` Runs the cURL HTTP client. The `-L` flag tells cURL to follow redirects (if any). *** **2. HTTP method** ```bash --request POST \ ``` Specifies that the request uses the **POST** method. *** **3. Endpoint** ```bash --url 'https://api.aimlapi.com/v1/chat/completions' \ ``` The full endpoint URL used to call chat models. *** **4. Authorization header** ```bash --header 'Authorization: Bearer ' \ ``` Sends your AIMLAPI key in the `Authorization` header. *** **5. Content type** ```bash --header 'Content-Type: application/json' \ ``` Indicates that the request body is JSON. *** **6. Request body** ```bash --data '{ "model": "google/gemma-3-4b-it", "messages": [ { "role": "user", "content": "Tell me about San Francisco" } ], "temperature": 0.7, "max_tokens": 512 }' ``` This is the payload sent to the API: * `model` – the model identifier. * `messages` – the chat history. * `role: "user"` – the user message. * `content` – the user prompt. * `temperature` – controls output randomness. * `max_tokens` – the maximum number of tokens in the response. {% hint style="info" %} These are the input parameters used to tell the endpoint—which in this case generates text answers—what exactly we want it to produce. With the parameters shown above, we are effectively asking the API to use the `google/gemma-3-4b-it` model and generate a reasonably vivid and engaging description of San Francisco, limited to roughly 300–350 words— with the `temperature` and `max_tokens` parameters controlling the creativity and approximate length of the output, respectively. {% endhint %} *** **7. Response** In the cURL example, you receive the **entire JSON response**.\ No fields are extracted — cURL simply prints the raw output.
JavaScript (Node.js) *** **1. Define the user prompt** ```js userPrompt = 'Tell me about San Francisco' ``` Stores the text of the user request. *** **2. Call the API** ```js const response = await fetch( 'https://api.aimlapi.com/v1/chat/completions', { ... } ); ``` Sends an HTTP request to the endpoint. *** **3. HTTP method** ```js method: 'POST', ``` Specifies that the request uses the **POST** method. *** **4. Headers** ```js headers: { 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, ``` * Sends your AIMLAPI key in the `Authorization` header. * Indicates that the request body is JSON. *** **5. Request body** ```js body: JSON.stringify({ model: 'google/gemma-3-4b-it', messages: [ { role: 'user', content: userPrompt } ], temperature: 0.7, max_tokens: 512, }), ``` This is the payload sent to the API: * `model` – the model identifier. * `messages` – the chat history. * `role: "user"` – the user message. * `content` – the user prompt. * `temperature` – controls output randomness. * `max_tokens` – the maximum number of tokens in the response. {% hint style="info" %} These are the input parameters used to tell the endpoint—which in this case generates text answers—what exactly we want it to produce. With the parameters shown above, we are effectively asking the API to use the `google/gemma-3-4b-it` model and generate a reasonably vivid and engaging description of San Francisco, limited to roughly 300–350 words— with the `temperature` and `max_tokens` parameters controlling the creativity and approximate length of the output, respectively. {% endhint %} *** **6. Parse the response** ```js const data = await response.json(); ``` Converts the API response into a JavaScript object. *** **7. Extract the model’s text output** ```js const answer = data.choices[0].message.content; ``` Reads the text of the first generated message. *** **8. Print the result** ```js console.log('User:', userPrompt); console.log('AI:', answer); ``` Output formatting: from the model’s full response, only the generated text is extracted, and it is presented together with the original prompt in a dialogue-style format. ***
Python *** **1. Import the HTTP library** ```python import requests ``` The `requests` library is used to send HTTP requests. *** **2. Define the user prompt** ```python user_prompt = "Tell me about San Francisco" ``` Stores the text of the user query. *** **3. Call the API** ```python response = requests.post( "https://api.aimlapi.com/v1/chat/completions", ... ) ``` Sends a POST request to the endpoint. *** **4. Headers** ```python headers={ "Authorization": "Bearer ", "Content-Type": "application/json" }, ``` * Sends your AIMLAPI key in the `Authorization` header. * Indicates that the request body is JSON. *** **5. Request body** ```python json={ "model": "google/gemma-3-4b-it", "messages": [ { "role": "user", "content": user_prompt } ], "temperature": 0.7, "max_tokens": 512, } ``` This is the payload sent to the API: * `model` – the model identifier. * `messages` – the chat history. * `role: "user"` – the user message. * `content` – the user prompt. * `temperature` – controls output randomness. * `max_tokens` – the maximum number of tokens in the response. {% hint style="info" %} These are the input parameters used to tell the endpoint—which in this case generates text answers—what exactly we want it to produce. With the parameters shown above, we are effectively asking the API to use the `google/gemma-3-4b-it` model and generate a reasonably vivid and engaging description of San Francisco, limited to roughly 300–350 words— with the `temperature` and `max_tokens` parameters controlling the creativity and approximate length of the output, respectively. {% endhint %} *** **6. Parse the response** ```python data = response.json() ``` Converts the JSON response into a Python dictionary. *** **7. Extract the model’s text output** ```python answer = data["choices"][0]["message"]["content"] ``` Reads the text of the first generated message. *** **8. Print the result** ```python print("User:", user_prompt) print("AI:", answer) ``` Output formatting: from the model’s full response, only the generated text is extracted, and it is presented together with the original prompt in a dialogue-style format.
*** ## Future Steps * [Move to production-ready models: see the guide for connecting GPT-4o](/quickstart/simple-model/requesting-more-advanced-models) * [Browse and compare AI models, including GPT, Claude, and many others, using the Playground](https://aimlapi.com/app/) * [Know more about supported SDKs](/quickstart/supported-sdks) * [Learn more about special text model capabilities](/capabilities/completion-or-chat-models) * [Join the community: get help and share your projects in our Discord](https://discord.com/invite/hvaUsJpVJf) # Requesting more advanced models This guide uses a more advanced model, [GPT-4o](/api-references/text-models-llm/openai/gpt-4o), and also explains how to use various chat model capabilities: * streaming mode * calling tools * uploading images to the model for analysis * uploading files to the model for analysis * web search {% hint style="info" %} If you need help with API keys or environment configuration, go back to the previous step and follow [the detailed quickstart guide](/quickstart/simple-model) for the Gemma 3 model. {% endhint %} *** ## Making an API Call The chat model used in this example is more advanced. In addition to regular user messages, it supports the `system` role in the `messages` parameter, which can be used to define global instructions that affect the model’s overall behavior, for example: ```python messages: [ { role: "system", content: "You are a travel agent. Be descriptive and helpful.", }, { role: "user", content: "Tell me about San Francisco", }, ], ``` Here’s the complete code you can use right away in a cURL, Python, or Node.js program. You only need to replace `` with your AIML API key from your account, provide your behavior instructions in the system prompt, and place your request to the model in the user prompt. {% tabs %} {% tab title="cURL" %} ```bash curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "gpt-4o", "messages": [ { "role": "system", "content": "You are a travel agent. Be descriptive and helpful.", }, { "role": "user", "content": "Tell me about San Francisco" } ], "temperature": 0.7, "max_tokens": 512 }' ``` {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript systemPrompt = 'You are a travel agent. Be descriptive and helpful.' // instructions userPrompt = 'Tell me about San Francisco' // your request async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'gpt-4o', messages:[ { role: 'system', content: systemPrompt, }, { role: 'user', content: userPrompt } ], temperature: 0.7, max_tokens: 512, }), }); const data = await response.json(); const answer = data.choices[0].message.content; console.log('User:', userPrompt); console.log('AI:', answer); } main(); ``` {% endcode %} {% endtab %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation system_prompt = "You are a travel agent. Be descriptive and helpful." user_prompt = "Tell me about San Francisco" response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"gpt-4o", "messages":[ { "role":"system", "content": system_prompt, }, { "role":"user", "content": user_prompt, } ], "temperature": 0.7, "max_tokens": 256, } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% endtabs %} *** ## **Using Streaming Mode** Streaming lets the model send partial responses as they’re generated instead of waiting for the full output — useful for real‑time feedback. ### Full Streaming Response (Raw Events) This example shows how to consume the streaming response as-is, without abstraction. Each chunk is processed in real time, exposing the full event structure returned by the API. Use this approach if you need: * access to all event types * fine-grained control over parsing * debugging or logging of raw responses * support for metadata beyond plain text {% tabs %} {% tab title="Python" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"gpt-4o", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ], "stream": True } ) # data = response.json() print(response.text) ``` {% endtab %} {% tab title="Python + OpenAI SDK" %} ```python from openai import OpenAI # Initialize the client client = OpenAI( # Insert your AIML API Key instead of : api_key="YOUR_AIMLAPI_KEY", base_url="https://api.aimlapi.com/v1" ) # Create a streaming chat completion stream = client.chat.completions.create( model="gpt-4o", messages=[ { "role": "user", "content": "Hi! What do you think about mankind?" } ], stream=True ) # Print raw chunks (similar to response.text in requests) for chunk in stream: print(chunk) ``` {% endtab %} {% endtabs %}
Example raw streaming response {% code overflow="wrap" expandable="true" %} ```json data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"RmYFV8ad65HP9F"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"Hello"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"fjE24R0ZOJr"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"!"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"qAlxZuNpvVvIIOm"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" As"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Zn3rsadkL8zHO"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" an"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"D1ss0WZmiGg8l"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" AI"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"bOHB8VYpq4G0W"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"OwZvgIyMlYVcIgH"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" I"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"u9lFaH3ngdK6MR"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" don't"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"KRFgmSe4yG"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" have"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"YL8zlQ9PjDF"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" personal"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Gzgb5OT"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" opinions"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Flz362J"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" or"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"XA0qqmSQr2jme"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" feelings"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"VA3dwaU"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"POplI0eiOWXpIPD"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" but"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"aDifMrQ8OH9i"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" I"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"ceVweUN2pByieS"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" can"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"txjYCds61AQp"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" provide"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"IGlSpZBf"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" an"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"BtPIfSvUXgRnl"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" overview"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"IYfRhEo"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" of"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"uh8pR2mNtYSNQ"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" various"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"ILZ0ffVW"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" perspectives"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Rgs"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" on"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"r7Awao2PSZ0DH"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" mankind"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"m8vJ3dzf"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"f2wZrEj0RqUFprg"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" Humanity"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"cCPi2qV"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" is"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"yNd7SUoXBojpA"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" often"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"VEaggK2dFS"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" viewed"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"8nhopBJZe"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" as"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"6xG2VkJLonAeF"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" a"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"WDu20GtJyN8Lep"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" complex"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"4bE4D3tS"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"DZtW3Ahopdgl"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" multif"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"8bS4GMzf3"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"aceted"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"ivtxUAov3l"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" species"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Xcq85kDt"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" capable"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"PfwZUtYS"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" of"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"DoyM4RGNLxnFc"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" remarkable"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"mUvVH"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" achievements"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"4fl"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"GUdfkDUkNBNO"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" profound"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"x4KCnLk"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" creativity"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"goTL4"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"gkqK9sezr258S93"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" People"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"c49BcmfXz"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" have"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Br7pbWtK86v"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" built"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"dzAoO36Siv"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" civilizations"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"yS"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"hiMiIGF7QM9BeJA"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" explored"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"IhuVoUB"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" space"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"qNqiO3hyXB"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"UVmzp6Y0qjb7Zkb"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"iiIw0gK2MP5D"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" developed"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"FJUJhv"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" technologies"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"pkQ"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" that"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"sAhx0IJoR0m"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" transform"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"YDTnhx"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" everyday"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"imFIYIz"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" life"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"6xJBjebVPfo"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"DKIPIwgAnVDj3g1"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" At"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"HhuMheG0mPcuI"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" the"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"yIQIWY1CXoW6"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" same"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"QcKwiqSqGRU"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" time"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"f6e6uGKikn5"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"1eXIFULDN1iS8b1"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" humans"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"GH0z8I36B"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" face"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"JLUmj9BN7PQ"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" significant"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"qdQg"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" challenges"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"KMzNb"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" such"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"8pw9I3FGElO"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" as"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"nY0RLEY6Am9zD"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" environmental"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"4r"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" degradation"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"1zGA"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"bhbZZCR7wNgWQkq"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" social"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"FcCsVIGji"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" inequalities"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"6kb"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":","},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Z4Zz2oDgc5zw0D6"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Q5XvheR2EWhq"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" geopolitical"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"ySW"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" conflicts"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"eiERwe"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"oNAsPbgeJSOuPMg"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" The"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"4TwzxlGRpebL"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" potential"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"lW3Jfo"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" for"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"Ejvws7kQryhN"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" both"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"HVm3EDKAkuA"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" positive"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"HMY8pYv"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" change"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"fbOaTSNWR"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"ETmTxHsFbCkw"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" destructive"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"WHk8"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" behavior"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"EvSYFf5"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" makes"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"yfwGRy20jz"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" mankind"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"vwJGC8sU"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" a"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"nHyqFYnTzVmVsE"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" subject"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"wtm8Wh9c"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" of"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"gnLF2uDFfg976"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" deep"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"BEc6wh2y2vV"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" contemplation"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"zf"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" and"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"vpg86EhZm5c3"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" varied"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"iWNJAcR7a"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":" viewpoints"},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"JRXUN"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{"content":"."},"logprobs":null,"finish_reason":null}],"usage":null,"obfuscation":"5yN6iGLyFLiQV0H"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[{"index":0,"delta":{},"logprobs":null,"finish_reason":"stop"}],"usage":null,"obfuscation":"4lkqbaPLDt"} data: {"id":"chatcmpl-DL2G9KdEY06xq8M0PqZ5rs5Jv13ok","object":"chat.completion.chunk","created":1773905945,"model":"gpt-4o-2024-08-06","service_tier":"default","system_fingerprint":"fp_f986a632b0","choices":[],"usage":{"prompt_tokens":16,"completion_tokens":102,"total_tokens":118,"prompt_tokens_details":{"cached_tokens":0,"audio_tokens":0},"completion_tokens_details":{"reasoning_tokens":0,"audio_tokens":0,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"obfuscation":"VChaI1ntRBrTy"} ``` {% endcode %}
### Streaming Response Processing (Text Extraction) This example shows how to process the streaming response to extract only the generated text. Instead of handling all event types, the code filters incoming chunks and prints the content as it arrives. Use this approach if you only need the generated text. {% tabs %} {% tab title="Python" %} {% code expandable="true" %} ```python import requests import json url = "https://api.aimlapi.com/v1/chat/completions" headers = { # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" } payload = { "model": "gpt-4o", "messages": [ {"role": "user", "content": "Explain quantum computing simply."} ], "stream": True } with requests.post(url, headers=headers, json=payload, stream=True) as r: # Iterate over the streaming response line by line for line in r.iter_lines(): if not line: continue # Skip empty lines # Decode bytes to string line = line.decode("utf-8") # SSE messages start with "data: " if not line.startswith("data: "): continue # Remove the "data: " prefix data_str = line[len("data: "):] # "[DONE]" indicates the end of the stream if data_str.strip() == "[DONE]": break try: # Parse JSON payload data = json.loads(data_str) except json.JSONDecodeError: continue # Skip malformed chunks # Ensure "choices" exists and is not empty choices = data.get("choices") if not choices: continue # Extract text delta (OpenAI-style streaming format) delta = data.get("choices", [{}])[0].get("delta", {}) content = delta.get("content") # Print text as it arrives if content: print(content, end="") ``` {% endcode %} {% endtab %} {% tab title="Python + OpenAI SDK" %} {% code expandable="true" %} ```python from openai import OpenAI client = OpenAI( # Insert your AIML API Key instead of : api_key="", base_url="https://api.aimlapi.com/v1" ) stream = client.chat.completions.create( model="gpt-4o", messages=[ {"role": "user", "content": "Explain quantum computing simply."} ], stream=True ) # Iterate over streaming chunks for chunk in stream: # Ensure choices exist and are not empty if not chunk.choices: continue delta = chunk.choices[0].delta content = getattr(delta, "content", None) # Print text as it arrives if content: print(content, end="") ``` {% endcode %} {% endtab %} {% endtabs %}
Example processed clean streaming response {% code overflow="wrap" %} ```json Quantum computing is a type of computing that uses principles of quantum mechanics to process information. Unlike classical computers, which use bits to represent data as 0s or 1s, quantum computers use quantum bits or qubits. Qubits have unique properties that give quantum computers more power in certain tasks: 1. **Superposition**: A qubit can exist in multiple states (i.e., both 0 and 1) simultaneously. This allows quantum computers to process a vast amount of possibilities at once. 2. **Entanglement**: Qubits can be linked together in such a way that the state of one qubit can depend on the state of another, no matter the distance apart. This can lead to more efficient processing and problem-solving. 3. **Quantum Interference**: Quantum algorithms make use of interference, where different quantum states can amplify or cancel each other out, guiding the computation toward the correct answer. Because of these properties, quantum computers have the potential to solve certain complex problems much faster than classical computers can, potentially revolutionizing fields like cryptography, materials science, and optimization. However, building practical quantum computers is extremely challenging due to issues with qubit stability and error rates. ``` {% endcode %}
*** ## **Tool calling** GPT‑4o can call functions/tools you define in the API request to extend behavior (e.g., performing calculations, retrieving structured data).
How it works 1. **Initial request** — The model receives the user prompt and the registered tool, and generates a `tool_calls` object indicating which function it wants to execute. 2. **Extract and run the tool** — Parse the `arguments` from the `tool_calls` object and execute the function locally. 3. **Send back the result** — Return the computed result to the model using the `tool` role and the `content` field. 4. **Final response** — The model incorporates the tool’s output and generates a complete answer for the user.
{% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # Insert your AIML API Key instead of : api_key = "" base_url = "https://api.aimlapi.com/v1" headers = { "Authorization": f"Bearer {api_key}", "Content-Type": "application/json" } # Step 1: Define the tool correctly tool = { "type": "function", "function": { "name": "toCelsius", "description": "Convert Fahrenheit to Celsius", "parameters": { "type": "object", "properties": { "fahrenheit": {"type": "number"} }, "required": ["fahrenheit"] } } } # Step 2: Initial request with the tool payload = { "model": "gpt-4o", "messages": [ {"role": "user", "content": "Convert 256°F to °C"} ], "tools": [tool] } response = requests.post(f"{base_url}/chat/completions", headers=headers, json=payload) data = response.json() # Step 3: Extract tool call tool_calls = data["choices"][0]["message"].get("tool_calls", []) if not tool_calls: raise ValueError("No tool calls found. Make sure the tool is correctly defined.") tool_call = tool_calls[0] arguments = json.loads(tool_call["function"]["arguments"]) fahrenheit = arguments["fahrenheit"] # Step 4: Execute the tool locally celsius_result = (fahrenheit - 32) * 5 / 9 # Step 5: Send result back to model final_payload = { "model": "gpt-4o", "messages": [ {"role": "user", "content": "Convert 256°F to °C"}, { "role": "assistant", "tool_calls": [ { "id": tool_call["id"], "type": "function", "function": { "name": tool_call["function"]["name"], "arguments": tool_call["function"]["arguments"] } } ] }, { "role": "tool", "tool_call_id": tool_call["id"], "content": str(celsius_result) } ] } final_response = requests.post(f"{base_url}/chat/completions", headers=headers, json=final_payload) final_data = final_response.json() # Step 6: Print final answer print(final_data["choices"][0]["message"]["content"]) ``` {% endcode %} {% endtab %} {% tab title="Python + OpenAI SDK" %} {% code overflow="wrap" %} ```python from openai import OpenAI import json client = OpenAI( # Insert your AIML API Key instead of : api_key="", base_url="https://api.aimlapi.com/v1" ) # Step 1: Define the tool correctly tool = { "type": "function", "function": { "name": "toCelsius", "description": "Convert Fahrenheit to Celsius", "parameters": { "type": "object", "properties": { "fahrenheit": {"type": "number"} }, "required": ["fahrenheit"] } } } # Step 2: Initial request with tool initial_response = client.chat.completions.create( model="gpt-4o", messages=[{"role": "user", "content": "Convert 256°F to °C"}], tools=[tool] ) # Step 3: Extract tool call assistant_message = initial_response.choices[0].message tool_calls = getattr(assistant_message, "tool_calls", []) if not tool_calls: raise ValueError("No tool calls found. Make sure the tool is correctly defined.") tool_call = tool_calls[0] arguments = json.loads(tool_call.function.arguments) fahrenheit = arguments["fahrenheit"] # Step 4: Execute tool locally celsius_result = (fahrenheit - 32) * 5 / 9 # Step 5: Send result back final_response = client.chat.completions.create( model="gpt-4o", messages=[ {"role": "user", "content": "Convert 256°F to °C"}, { "role": "assistant", "tool_calls": [ { "id": tool_call.id, "type": "function", "function": { "name": tool_call.function.name, "arguments": tool_call.function.arguments, }, } ], }, { "role": "tool", "tool_call_id": tool_call.id, "content": str(celsius_result), }, ], ) print(final_response.choices[0].message.content) ``` {% endcode %} {% endtab %} {% endtabs %}
Example response {% code overflow="wrap" expandable="true" %} ```json 256°F is approximately 124.44°C. ``` {% endcode %}
*** ## **Image upload** GPT‑4o supports vision inputs: you can send an image URL in the `messages` request to let the model analyze or describe it. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json url = "https://api.aimlapi.com/v1/chat/completions" headers = { # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" } payload = { "model": "gpt-4o", "messages": [ { "role": "user", "content": [ {"type": "text", "text": "Describe this scene:"}, {"type": "image_url", "image_url": {"url": "https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/mona_lisa_extended.jpg"}} ] } ] } response = requests.post(url, headers=headers, data=json.dumps(payload)) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="Python + OpenAI SDK" %} {% code overflow="wrap" %} ```python from openai import OpenAI import json # Initialize the client client = OpenAI( # Insert your AIML API Key instead of : api_key="", base_url="https://api.aimlapi.com/v1" ) # Prepare the messages with text and image_url messages = [ { "role": "user", "content": [ {"type": "text", "text": "Describe this scene:"}, { "type": "image_url", "image_url": { "url": "https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/mona_lisa_extended.jpg" } } ] } ] # Create a chat completion response = client.chat.completions.create( model="gpt-4o", messages=messages ) # Print full JSON response print(json.dumps(response.model_dump(), indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% endtabs %}
Example response {% code overflow="wrap" %} ```json { "id": "chatcmpl-DL3DDPif2s79HbOHySq6bVY8SAsKQ", "choices": [ { "finish_reason": "stop", "index": 0, "logprobs": null, "message": { "content": "The scene is an iconic Renaissance portrait showing a woman with an enigmatic smile, known for its mastery of detail and composition. The woman is seated against a distant, dreamlike landscape featuring winding paths and rocky formations. She wears a dark dress and light veil, with her hands delicately folded. The background's atmospheric perspective creates depth, with bluish mountains fading into the horizon. The artwork evokes a sense of mystery and balance.", "refusal": null, "role": "assistant", "annotations": [], "audio": null, "function_call": null, "tool_calls": null } } ], "created": 1773909607, "model": "gpt-4o-2024-08-06", "object": "chat.completion", "service_tier": "default", "system_fingerprint": "fp_0a8aa8bfeb", "usage": { "completion_tokens": 85, "prompt_tokens": 776, "total_tokens": 861, "completion_tokens_details": { "accepted_prediction_tokens": 0, "audio_tokens": 0, "reasoning_tokens": 0, "rejected_prediction_tokens": 0 }, "prompt_tokens_details": { "audio_tokens": 0, "cached_tokens": 0 } }, "meta": { "usage": { "credits_used": 7254 } } } ``` {% endcode %}
*** ## **Web search integration** With *search‑preview* models, you can perform live web search queries in combination with the model to get up‑to‑date results and grounded responses. {% hint style="info" %} See [the complete list of our search‑capable models](/capabilities/web-search#models-that-support-web-search). {% endhint %} {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import json import requests from typing import Dict, Any # Insert your AIML API Key instead of : API_KEY = "" BASE_URL = "https://api.aimlapi.com/v1" HEADERS = { "Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json", } def search_impl(arguments: Dict[str, Any]) -> Any: return arguments def chat(messages): url = f"{BASE_URL}/chat/completions" payload = { "model": "gpt-4o-mini-search-preview", "messages": messages, "temperature": 0.6, "tools": [ { "type": "builtin_function", "function": {"name": "$web_search"}, } ] } response = requests.post(url, headers=HEADERS, json=payload) response.raise_for_status() return response.json()["choices"][0] def main(): messages = [ {"role": "system", "content": "You are GPT with web search skills."}, {"role": "user", "content": "Please search for AGI and tell me what it is in English."} ] finish_reason = None while finish_reason is None or finish_reason == "tool_calls": choice = chat(messages) finish_reason = choice["finish_reason"] message = choice["message"] if finish_reason == "tool_calls": messages.append(message) for tool_call in message["tool_calls"]: tool_call_name = tool_call["function"]["name"] tool_call_arguments = json.loads(tool_call["function"]["arguments"]) if tool_call_name == "$web_search": tool_result = search_impl(tool_call_arguments) else: tool_result = f"Error: unable to find tool by name '{tool_call_name}'" messages.append({ "role": "tool", "tool_call_id": tool_call["id"], "name": tool_call_name, "content": json.dumps(tool_result), }) print(message["content"]) if __name__ == "__main__": main() ``` {% endcode %} {% endtab %} {% tab title="Python + OpenAI SDK" %} {% code overflow="wrap" %} ```python import json from typing import Dict, Any from openai import OpenAI # Insert your API key client = OpenAI( # Insert your AIML API Key instead of : api_key="YOUR_AIMLAPI_KEY", base_url="https://api.aimlapi.com/v1" ) def search_impl(arguments: Dict[str, Any]) -> Any: return arguments def chat(messages): response = client.chat.completions.create( model="gpt-4o-mini-search-preview", messages=messages, temperature=0.6, tools=[ { "type": "function", "function": { "name": "$web_search", "parameters": { "type": "object", "properties": {}, }, }, } ], ) return response.choices[0] def main(): messages = [ {"role": "system", "content": "You are GPT with web search skills."}, {"role": "user", "content": "Please search for AGI and tell me what it is in English."} ] finish_reason = None while finish_reason is None or finish_reason == "tool_calls": choice = chat(messages) finish_reason = choice.finish_reason message = choice.message if finish_reason == "tool_calls": messages.append(message.model_dump()) for tool_call in message.tool_calls: tool_call_name = tool_call.function.name tool_call_arguments = json.loads(tool_call.function.arguments) if tool_call_name == "$web_search": tool_result = search_impl(tool_call_arguments) else: tool_result = f"Error: unable to find tool by name '{tool_call_name}'" messages.append({ "role": "tool", "tool_call_id": tool_call.id, "name": tool_call_name, "content": json.dumps(tool_result), }) print(message.content) if __name__ == "__main__": main() ``` {% endcode %} {% endtab %} {% endtabs %}
Example response {% code overflow="wrap" %} ```json "AGI" is an acronym that can represent different terms depending on the context: 1. **Adjusted Gross Income**: In the United States, AGI refers to Adjusted Gross Income, which is a taxpayer's total income from all sources minus allowable adjustments. This figure is used to determine taxable income and eligibility for various tax benefits. ([usafacts.org](https://usafacts.org/articles/adjusted-gross-income-agi-definition?utm_source=openai)) 2. **Artificial General Intelligence**: In the field of artificial intelligence, AGI stands for Artificial General Intelligence. This concept refers to AI systems that possess the ability to understand, learn, and apply knowledge across a wide range of tasks, matching or surpassing human cognitive abilities. ([en.wikipedia.org](https://en.wikipedia.org/wiki/Artificial_general_intelligence?utm_source=openai)) 3. **Alliance Graphique Internationale**: AGI also denotes the Alliance Graphique Internationale, an international organization of leading graphic artists and designers. ([en.wikipedia.org](https://en.wikipedia.org/wiki/Alliance_Graphique_Internationale?utm_source=openai)) 4. **Agi Language**: Additionally, "Agi" is the name of a Torricelli language spoken in Papua New Guinea. ([en.wikipedia.org](https://en.wikipedia.org/wiki/Agi_language?utm_source=openai)) The specific meaning of "AGI" depends on the context in which it is used. ``` {% endcode %}
*** ## Future Steps * [Browse and compare AI models, including GPT, Claude, and many others, using the Playground](https://aimlapi.com/app/) * [Know more about supported SDKs](/quickstart/supported-sdks) * [Learn more about special text model capabilities](/capabilities/completion-or-chat-models) * [Join the community: get help and share your projects in our Discord](https://discord.com/invite/hvaUsJpVJf) # MCP Connect AI/ML API to your AI client — **Claude Desktop, Claude (web), Cursor, Claude Code**, and other MCP-capable apps — over the [Model Context Protocol](https://modelcontextprotocol.io/) using **OAuth**. You sign in with your AI/ML API account in the browser; there is **no API key to copy or paste**. Once connected, you can discover and compare models, run inference (LLM / image / video / audio / embeddings), manage long-running generation jobs, and check your balance — directly from the client. Usage is billed to your AI/ML API account, exactly like the REST API. > Prefer to paste an API key instead of signing in? That method is also supported — add an `Authorization: Bearer ` header instead of using OAuth. This guide covers the **OAuth** (browser sign-in) flow. ## Prerequisites * An **AI/ML API account** — create one at [https://aimlapi.com](https://aimlapi.com/). * A client that supports **remote MCP servers over Streamable HTTP with OAuth** (all clients below do). * A balance on your account if you plan to run billable inference (top up at ). * **A correct system clock.** OAuth tokens are time-sensitive — if your computer's date, time, or time zone is wrong, the sign-in will fail (and some clients won't even start the session). Make sure the clock is set automatically / synced before you begin. ## Endpoint | Environment | MCP URL | | ----------- | ----------------------------- | | Production | `https://mcp.aimlapi.com/mcp` | You only ever enter this **one URL**. Your client discovers everything else (the sign-in page, how to register, how to get a token) automatically. ## How the OAuth connection works When you add the server URL **without an API key**, your client: 1. Contacts the MCP server and is told that authentication is required. 2. Automatically discovers the AI/ML API authorization server. 3. Registers itself and opens your browser. 4. You **sign in to AI/ML API** and **approve access**; the client receives a token and connects. You never handle a key. The client stores the token securely and refreshes it for you. > **Note on the "Connect" button (Claude Desktop & web).** Adding the connector only *registers* it — it does not sign you in. After you add it, the connector appears with a **Connect** button; you must click **Connect** to open the browser and complete the AI/ML API sign-in. Until you click **Connect** and approve, the connector stays added but **not authenticated**, and its tools will not work. ## Claude Desktop 1. **Settings → Connectors → Add custom connector.** 2. **Name:** `AIMLAPI` · **Remote MCP server URL:** `https://mcp.aimlapi.com/mcp` 3. Click **Add**. The connector is now registered. 4. Click **Connect** on the AIMLAPI connector. A browser window opens → **sign in to AI/ML API** → **approve access**. 5. The connector shows **connected**, and the AI/ML API tools become available in your chats. 6. **Verify:** open a chat and ask *"Use AI/ML API to check my account balance."* A returned balance confirms the connection and authorization are working. **Remove:** **Settings → Connectors → AI/ML API → Remove / Disconnect.** ## Claude (web — claude.ai) 1. Open **Settings → Connectors → Add custom connector** (on Team/Enterprise plans this is done by an Owner under **Organization settings → Connectors**). 2. Enter the URL `https://mcp.aimlapi.com/mcp` and add it. 3. Click **Connect** on the AIMLAPI connector. A browser flow opens → **sign in to AI/ML API** → **approve** → connected. 4. **Verify:** ask *"Use AI/ML API to check my account balance."* > Custom-connector availability depends on your Claude plan. **Remove:** **Settings → Connectors → AI/ML API → Remove.** ## Cursor Add the server to `~/.cursor/mcp.json` (global) or `.cursor/mcp.json` (project). **Do not add a `headers` block** — leaving it out is what triggers the OAuth flow: ```json { "mcpServers": { "aimlapi": { "url": "https://mcp.aimlapi.com/mcp" } } } ``` Then open **Cursor Settings → Tools & Integrations** and find **aimlapi**. It usually appears **Disabled** first — click **···→ Enable**, then click **Login**. A browser opens → **sign in to AI/ML API** → **Allow**. The server turns green when connected. **Verify:** in an Agent chat, ask *"Using aimlapi, check my account balance."* **Remove:** Cursor cannot delete an MCP server from the UI — the **···** menu only offers Configure / Reload / Enable. To remove it, open `~/.cursor/mcp.json` (the **Configure** dialog shows its path) and delete the `aimlapi` block, then **Reload** or restart Cursor. Note: turning off the source toggle only hides the server from the UI — the entry stays in the file until you delete it there. ## Claude Code (CLI) ```bash claude mcp add --transport http aimlapi https://mcp.aimlapi.com/mcp ``` By default this adds the server in **local** scope — it is only available inside the current folder / project. To make it available everywhere, add `--scope user`: ```bash claude mcp add --transport http --scope user aimlapi https://mcp.aimlapi.com/mcp ``` Then, inside an interactive Claude Code session, run **`/mcp`**, select **aimlapi**, and choose **Authenticate** — a browser opens for you to sign in. (Do **not** pass an `--header` flag; that switches the client to API-key auth instead of OAuth.) **Verify:** run `claude mcp list` (the server should show as connected), or in a session ask *"Use the aimlapi MCP to check my account balance."* **Remove:** `claude mcp remove aimlapi` (confirm with `claude mcp list`). ## Other MCP clients Any client implementing the MCP authorization spec (OAuth 2.1 + PKCE with dynamic client registration) works the same way: point it at `https://mcp.aimlapi.com/mcp` with no API key and complete the browser sign-in. To inspect the flow step by step, use the MCP Inspector: ```bash npx @modelcontextprotocol/inspector@latest ``` Set **Transport = Streamable HTTP**, **URL = `https://mcp.aimlapi.com/mcp`**, then use **Open Auth Settings → Quick OAuth Flow**. ## What you can do once connected * **Discover models** — search and compare the catalog, get a comparable headline price. * **Run inference** — LLM chat, embeddings, image / video / audio generation. * **Manage jobs** — submit long-running generations and poll their status/results. * **Account** — check your balance and usage. Every call is billed to your AIML API account, just like the REST API. ### Example requests Once connected, you talk to AIML API in plain language from your client. Because AIML API is an aggregator of **1000+ models**, you can do things no single-provider MCP can. **Pick the right model for the job** *(unique to an aggregator)* * "Find the cheapest model with function calling and 128k context, and show the cost per 1M tokens." * "Compare GPT-5.6 Sol and Claude Opus 4.8 on price and context window." * "Which AIML API model is best for coding agents?" * "Run this prompt through the top 3 models and show me the answers side by side." **Keep an eye on spend (AI FinOps)** * "What's my AIML API balance?" * "How much have I spent this month, and on what?" * "Suggest a cheaper model for my last task." **Discover the catalog** * "List the newest LLMs on AIML API." * "Show me all image models and their prices." **Run inference** * "Explain OAuth in two sentences." * "Generate an image of a mountain lake at sunset with Nano Banana 2." * "Create embeddings for these three sentences." Your client picks the right AIML API tool automatically and streams the result back into the chat. ## Managing the connection To stop using AIML API from a client, **remove or disconnect** the AI/ML API connector — the exact step differs per client: | Client | How to remove | | -------------- | ----------------------------------------------------------------------------------------------------------------------- | | Claude Desktop | Settings → Connectors → AI/ML API → Remove / Disconnect. | | Claude (web) | Settings → Connectors → AI/ML API → Remove. | | Cursor | Edit `~/.cursor/mcp.json` and delete the `aimlapi` block (cannot be removed from the UI), then Reload / restart Cursor. | | Claude Code | `claude mcp remove aimlapi`. | **Clean reinstall.** Removing the connector in the client does not always revoke the OAuth token on the AI/ML API side. For a fully clean reinstall — one that shows the sign-in / approval screen again — also revoke the app's access in your AI/ML API account (**Authorized apps / Connections → Revoke**) before adding the server back. ## Troubleshooting
SymptomFix
Browser didn't open, or the client is stuck "connecting"Remove the connector and add it again; make sure you clicked Connect and completed the AI/ML API sign-in in the browser.
Session won't start / "clock is set incorrectly" / sign-in fails immediatelyYour system clock is wrong. Set the date, time, and time zone automatically, then retry — OAuth tokens are time-sensitive.
Connector added but tools don't workYou likely didn't click Connect (Desktop / web) or Enable + Login (Cursor). The connector must be authenticated, not just added.
Server shows Disabled in Cursor after editing mcp.jsonOpen ···→ Enable, then Login to authenticate.
MCP works in one folder but is missing in another (Claude Code)It was added in local scope. Re-add with --scope user to make it global.
Reinstalled but never asked to sign in againThe old OAuth token is still valid. Revoke the app in AI/ML API → Authorized apps, then reconnect.
Connected, but inference fails with an insufficient-balance errorTop up at https://aimlapi.com/app/billing.
The client connects but never asks you to sign inYou likely added an Authorization header — with a header the client uses API-key auth, not OAuth. Remove the header to use OAuth.
"Sign-in failed" / authorization errorConfirm you're signing in to the correct AI/ML API account, then retry from the client.
# Supported SDKs A description of the software development kits (SDKs) that can be used to interact with the AIML API. This page describes the SDK[^1]s that can be used to call our API.
Key Definitions & Notes The **REST API** itself is *not* an SDK. It is the **server-side interface** that exposes your models over HTTP. It defines endpoints, HTTP methods (POST/GET), required headers, and the structure of request and response JSON. Essentially, it’s the “contract” the server provides for clients to interact with models programmatically. *** An **SDK** (Software Development Kit) is a **client-side library** that wraps around the REST API. It handles details like building HTTP requests, serializing/deserializing JSON, error handling, retries, and sometimes additional conveniences. You can skip the SDK and call the REST API directly via cURL, fetch, requests, etc.\ The SDK just makes your life easier; the REST API is the “core interface” the SDK talks to. *** The following flow shows how a request travels from your code to the model and back. Using an SDK is optional — it simply wraps the REST API for convenience. > Your code → SDK (optional) → REST API → Model → REST API → SDK → Your code
{% hint style="info" %} Comparing requests made with raw REST API and different SDKs, pay attention to the following common aspects: * how the Authorization header and the AIML API key are provided, * how the `POST` method and the endpoint URL are specified, * how the input parameters are passed. {% endhint %} {% hint style="success" %} Also take a look at the [**INTEGRATIONS**](/integrations/our-integration-list) section — it covers many third-party services and libraries (workflow platforms, coding assistants, etc.) that allow you to integrate our models in various ways. {% endhint %} *** ## REST API We use the REST API because it’s fast, simple, and easy to understand. Only in Python do you need to import a separate library (`requests`), while cURL and JavaScript (Node.js) already have built-in support for HTTP requests. Therefore, REST API is used in the documentation examples for *all* of our models. ### Installation In Python examples, you need to import the `requests` library. The Node.js and cURL examples do not require any additional imports. Install the library first: {% tabs %} {% tab title="Shell" %} ```bash pip install requests ``` {% endtab %} {% endtabs %} Import the library in every Python code snippet where you make calls to the REST API. {% tabs %} {% tab title="Python" %} ```python import requests ``` {% endtab %} {% endtabs %} ### Authorization Our API authorization is based on a Bearer token. Include it in the `Authorization` HTTP header within the request. Example: {% tabs %} {% tab title="cURL" %} ```ruby --header 'Authorization: Bearer ' \ ``` {% endtab %} {% tab title="JavaScript" %} ```javascript headers: { Authorization: "Bearer ", }, ``` {% endtab %} {% tab title="Python" %} ```python headers={ "Authorization": "Bearer ", }, ``` {% endtab %} {% endtabs %} ### Request Example {% tabs %} {% tab title="cURL" %} ```ruby curl --request POST \ --url https://api.aimlapi.com/chat/completions \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "google/gemma-3-4b-it", "messages": [ { "role": "user", "content": "What kind of model are you?" } ], "max_tokens": 512 }' ``` {% endtab %} {% tab title="JavaScript" %} ```javascript fetch("https://api.aimlapi.com/chat/completions", { method: "POST", headers: { Authorization: "Bearer ", "Content-Type": "application/json", }, body: JSON.stringify({ model: "google/gemma-3-4b-it", messages: [ { role: "user", content: "What kind of model are you?", }, ], max_tokens: 512, }), }) .then((res) => res.json()) .then(console.log); ``` {% endtab %} {% tab title="Python" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( url="https://api.aimlapi.com/chat/completions", headers={ "Authorization": "Bearer ", "Content-Type": "application/json", }, data=json.dumps( { "model": "google/gemma-3-4b-it", "messages": [ { "role": "user", "content": "What kind of model are you?", }, ], "max_tokens": 512 } ), ) response.raise_for_status() print(response.json()) ``` {% endtab %} {% endtabs %} *** ## OpenAI The OpenAI SDK is a convenient library that simplifies working with our API. It automatically handles JSON responses, includes built-in error handling and retry logic, and provides simple, easy-to-use methods for all API features such as chat, embeddings, and completions.
The AI features that the OpenAI SDK supports * Streaming * Completions * Chat Completions * Audio * Beta Assistants * Beta Threads * Embeddings * Image Generation * File Uploads {% hint style="info" %} Therefore, we don’t currently have the option to call video models or voice / speech models (STT and TTS) through this SDK. {% endhint %}
### Installation
Python 1\. Make sure you have Python 3.7+ and `pip` installed. 2\. Install the OpenAI SDK via terminal or Jupyter Notebook: ```bash pip install openai ``` > In Jupyter Notebook, you can also use: > > ```python > %pip install openai > ``` 3\. Import the SDK: ```python import openai ``` ***
JavaScript (Node.js) 1\. Make sure you have Node.js 18+ and npm installed. 2\. Install the OpenAI SDK in your project: ```bash npm install openai ``` 3\. Import the SDK and initialize the client: ```js import OpenAI from "openai"; ```
### Example Code {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python from openai import OpenAI # Insert your AIML API key in the quotation marks instead of : api_key = "" base_url = "https://api.aimlapi.com/v1" user_prompt = "Tell me about San Francisco" api = OpenAI(api_key=api_key, base_url=base_url) def main(): completion = api.chat.completions.create( model="google/gemma-3-4b-it", messages=[ { "role": "user", "content": user_prompt }, ], temperature=0.7, max_tokens=256, ) response = completion.choices[0].message.content print("User:", user_prompt) print("AI:", response) if __name__ == "__main__": main() ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript #!/usr/bin/env node const OpenAI = require("openai"); const baseURL = "https://api.aimlapi.com/v1"; const apiKey = ""; const systemPrompt = "You are a travel agent. Be descriptive and helpful."; const userPrompt = "Tell me about San Francisco"; const api = new OpenAI({ apiKey, baseURL, }); const main = async () => { try { const completion = await api.chat.completions.create({ model: "gpt-4o", messages: [ { role: "system", content: systemPrompt, }, { role: "user", content: userPrompt, }, ], temperature: 0.7, max_tokens: 256, }); const response = completion.choices[0].message.content; console.log("User:", userPrompt); console.log("AI:", response); } catch (error) { console.error("Error:", error.message); } }; main(); ``` {% endcode %} {% endtab %} {% endtabs %} *** ## AI/ML API Python library We have started developing our own SDK to simplify the use of our service. {% hint style="success" %} If you’d like to contribute to expanding its functionality, feel free to reach out to us on [**Discord**](https://discord.com/invite/hvaUsJpVJf)! {% endhint %} ### Installation After obtaining your AIML API key, create an .env file and copy the required contents into it. ```shell touch .env ``` Copy the code below, paste it into your `.env` file, and set your API key in `AIML_API_KEY=""`, replacing `` with your actual key: ```shell AIML_API_KEY = "" AIML_API_URL = "https://api.aimlapi.com/v1" ``` Install the package: ```shell # install from PyPI pip install aimlapi-sdk-python ``` ### Request Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python from aiml_api import AIML_API api = AIML_API() completion = api.chat.completions.create( model = "mistralai/Mistral-7B-Instruct-v0.2", messages = [ {"role": "user", "content": "Explain the importance of low-latency LLMs"}, ], temperature = 0.7, max_tokens = 256, ) response = completion.choices[0].message.content print("AI:", response) ``` {% endcode %} {% endtab %} {% endtabs %} To execute the script, use: ```shell python3 .py ``` *** ## Next Steps * [Check our full list of model IDs](/api-references/model-database) * [Browse and compare AI models, including GPT, Claude, and many others, using the Playground](https://aimlapi.com/app/) * [Learn more about special text model capabilities](/capabilities/completion-or-chat-models) * [Join the community: get help and share your projects in our Discord](https://discord.com/invite/hvaUsJpVJf) [^1]: An SDK (Software Development Kit) is a **client-side library** that wraps around the REST API. # Service Endpoints # Account Balance ## Get account balance info {% hint style="warning" %} This endpoint is considered legacy and is scheduled for future deprecation.\ Please plan to migrate to the new `/v2/billing` and `/v2/billing/detail` endpoints documented below. {% endhint %} You can query your account balance and other billing details through this API.\ To make a request, you only need your AIMLAPI key obtained from your [account dashboard](https://aimlapi.com/app/keys). ## GET /v1/billing/balance > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/billing/balance":{"get":{"operationId":"_v1_billing_balance","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"–","title":"–"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"balance":{"type":"number","description":"The total credits associated with the provided API key."},"lowBalance":{"type":"boolean","description":"True if the balance is below the threshold."},"lowBalanceThreshold":{"type":"number","description":"Threshold for switching to low balance status."},"lastUpdated":{"type":"string","format":"date-time","description":"The date of the request — i.e., the current date."},"autoDebitStatus":{"type":"string","description":"Indicates whether auto top-up is enabled for the plan."},"status":{"type":"string","description":"The status of the plan associated with the provided API key."},"statusExplanation":{"type":"string","description":"A more detailed explanation of the plan status."}},"required":["balance","lowBalance","lowBalanceThreshold","lastUpdated","autoDebitStatus","status","statusExplanation"]}}}}}}}}} ``` ## Get balance info Returns a user's balance. ## GET /v2/billing > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v2/billing":{"get":{"operationId":"_v2_billing","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"–","title":"–"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"current_balance":{"type":"number","description":"current user balance in USD."},"currency":{"type":"string","description":"balance currency (always USD)"}},"required":["current_balance","currency"]}}}}}}}}} ``` ## Get detailed billing info Returns detailed billing information, balance and auto top-up settings. ## GET /v2/billing/detail > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v2/billing/detail":{"get":{"operationId":"_v2_billing_detail","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"–","title":"–"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"user_id":{"type":"number","description":"User ID."},"email":{"type":"string","description":"User email."},"current_balance":{"type":"number","description":"Current balance in USD."},"currency":{"type":"string","description":"Currency (always USD)."},"autotopup_settings":{"type":"object","description":"Auto top-up settings.","properties":{"is_enabled":{"type":"boolean","description":"Whether auto top-up is enabled."},"threshold":{"type":"number","description":"Balance threshold that triggers auto top-up (USD)."},"amount":{"type":"number","description":"Auto top-up amount (USD)."},"currency":{"type":"string","description":"Auto top-up currency (always USD)."}},"required":["user_id","email","current_balance","currency","autotopup_settings"]}}}}}}}}}}} ``` # API Key Management
Create a new API keyPOST https://api.aimlapi.com/v1/keys
List API keysGET https://api.aimlapi.com/v1/keys
Get the API keyGET https://api.aimlapi.com/v1/key
Update an API keyPATCH https://api.aimlapi.com/v1/keys/{prefix}
Delete an API keyDELETE https://api.aimlapi.com/v1/keys/{prefix}
{% hint style="warning" %} Before you start you should create [a management key](/faq/how-can-i-work-with-my-api-keys#management-keys). {% endhint %} ### Create a new API key Copy the created key and store it in a secure location. If the key is lost, create a new one. ## POST /v1/keys > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/keys":{"post":{"operationId":"KeysController_createKey_v1","requestBody":{"required":false,"content":{"application/json":{"schema":{"type":"object","properties":{"name":{"type":"string","description":"Optional human-readable name of the API key."},"limit":{"type":"object","description":"Optional spending limit configuration.","properties":{"retention":{"type":"string","description":"Limit period.","enum":["no_reset","day","week","month"]},"threshold":{"type":"number","description":"Spending limit threshold for the selected period, in USD."}}},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}}},"additionalProperties":false}}}},"responses":{"200":{"description":"API key creation result","content":{"application/json":{"schema":{"type":"object","properties":{"data":{"type":"object","properties":{"name":{"type":"string","nullable":true,"description":"Human-readable, user-defined name for the API key."},"disabled":{"type":"boolean","description":"Indicates whether the key is disabled."},"prefix":{"type":"string","description":"Key prefix. This is the first 8 characters of your API key, visible in the dashboard."},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}},"limit":{"type":"object","description":"Applied spending limits.","properties":{"retention":{"type":"string","description":"Limit period.","enum":["no_reset","day","week","month"]},"threshold":{"type":"number","description":"Spending limit threshold for the selected period, in USD"}}},"created_at":{"type":"string","format":"date-time","description":"Creation timestamp (UTC)."},"updated_at":{"type":"string","format":"date-time","description":"Last update timestamp (UTC)."},"monthly_usage":{"type":"number","description":"Current monthly usage amount."},"key":{"type":"string","description":"Full API key value (returned only at creation time)."}},"required":["disabled","prefix","scopes","created_at","updated_at","monthly_usage","key"]}},"required":["data"]}}}}}}}}} ``` *** ### List API keys Returns all API keys for your account, including each key’s settings and metadata. ## GET /v1/keys > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/keys":{"get":{"operationId":"KeysController_listKeys_v1","responses":{"200":{"description":"List of API keys, ordered from oldest to newest","content":{"application/json":{"schema":{"type":"object","properties":{"data":{"type":"array","items":{"type":"object","properties":{"name":{"type":"string","nullable":true,"description":"Human-readable, user-defined name for the API key."},"prefix":{"type":"string","description":"Key prefix. This is the first 8 characters of your API key, visible in the dashboard. You can also obtain this value via the POST method (see the `prefix` field in its response)."},"disabled":{"type":"boolean","description":"Indicates whether the key is disabled."},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}},"limit":{"nullable":true,"type":"object","description":"Spending limits configuration (if set).","properties":{"retention":{"type":"string","enum":["no_reset","day","week","month"]},"threshold":{"type":"number","description":"Spending limit threshold for the selected period, in USD."}}},"created_at":{"type":"string","format":"date-time","description":"Creation timestamp (UTC)."},"updated_at":{"type":"string","format":"date-time","description":"Last update timestamp (UTC)."},"monthly_usage":{"type":"number","description":"Current monthly usage amount."}},"required":["disabled","prefix","scopes","created_at","updated_at","monthly_usage"]}}},"required":["data"]}}}}}}}}} ``` *** ### Get the API key Retrieve parameters of the most recently used API key. ## GET /v1/key > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/key":{"get":{"operationId":"KeysController_getLatestKey_v1","responses":{"200":{"description":"Parameters of the latest API key","content":{"application/json":{"schema":{"type":"object","properties":{"data":{"type":"object","properties":{"name":{"type":"string","nullable":true,"description":"Human-readable, user-defined name for the API key."},"disabled":{"type":"boolean","description":"Indicates whether the key is disabled."},"prefix":{"type":"string","description":"Key prefix. This is the first 8 characters of your API key, visible in the dashboard. You can also obtain this value via the POST method (see the `prefix` field in its response)."},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}},"limit":{"nullable":true,"type":"object","description":"Spending limits configuration (if set).","properties":{"retention":{"type":"string","enum":["no_reset","day","week","month"]},"threshold":{"type":"number","description":"Spending limit threshold for the selected period, in USD."}}},"created_at":{"type":"string","format":"date-time","description":"Creation timestamp (UTC)."},"updated_at":{"type":"string","format":"date-time","description":"Last update timestamp (UTC)."},"monthly_usage":{"type":"number","description":"Current monthly usage amount."}},"required":["disabled","prefix","created_at","updated_at","monthly_usage"]}},"required":["data"]}}}}}}}}} ``` *** ### Update an API key ## PATCH /v1/keys/{prefix} > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/keys/{prefix}":{"patch":{"operationId":"KeysController_patchKey_v1","parameters":[{"name":"prefix","in":"path","required":true,"description":"Prefix of the API key to update. Passed in the URL path. This is the first 8 characters of your API key, visible in the dashboard. You can also obtain this value via the GET method (see the `prefix` field in its response).","schema":{"type":"string"}}],"requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"name":{"type":"string","description":"Optional human-readable name of the API key."},"disabled":{"type":"boolean","description":"Enable or disable the API key."},"limit":{"type":"object","description":"Optional spending limit configuration.","properties":{"retention":{"type":"string","description":"Limit period.","enum":["no_reset","day","week","month"]},"threshold":{"type":"number","description":"Spending limit threshold for the selected period, in USD"}}},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}}}}}}},"responses":{"200":{"description":"Updated API key parameters","content":{"application/json":{"schema":{"type":"object","properties":{"data":{"type":"object","properties":{"name":{"type":"string","nullable":true,"description":"Human-readable, user-defined name for the API key."},"disabled":{"type":"boolean","description":"Indicates whether the key is disabled."},"prefix":{"type":"string","description":"Key prefix. This is the first 8 characters of your API key, visible in the dashboard. You can also obtain this value via the GET method (see the `prefix` field in its response)."},"scopes":{"type":"array","nullable":true,"description":"List of model access scopes assigned to the key. Defines which categories of models can be called using this API key.","items":{"type":"string","enum":["model:chat","model:responses","model:image","model:audio","model:video","model:embeddings","model:speech","model:ocr"]}},"limit":{"nullable":true,"type":"object","description":"Spending limits configuration (if set).","properties":{"retention":{"type":"string","enum":["no_reset","day","week","month"]},"threshold":{"type":"number"}}},"created_at":{"type":"string","format":"date-time","description":"Creation timestamp (UTC)."},"updated_at":{"type":"string","format":"date-time","description":"Last update timestamp (UTC)."},"monthly_usage":{"type":"number","description":"Current monthly usage amount."}},"required":["disabled","prefix","scopes","created_at","updated_at","monthly_usage"]}},"required":["data"]}}}}}}}}} ``` *** ### Delete an API key ## DELETE /v1/keys/{prefix} > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/keys/{prefix}":{"delete":{"operationId":"KeysController_postKeys_v1","parameters":[{"name":"prefix","in":"path","required":true,"description":"Prefix of the API key to delete. Passed in the URL path. This is the first 8 characters of the API key you want to delete. Passed in the URL path. This is the first 8 characters of your API key, visible in the dashboard. You can also obtain this value via the GET method (see the `prefix` field in its response).","schema":{"type":"string"}}],"responses":{"200":{"description":"Key deletion result","content":{"application/json":{"schema":{"type":"object","properties":{"data":{"type":"object","properties":{"prefix":{"type":"string","description":"Prefix of the deleted API key."},"deleted":{"type":"boolean","description":"Indicates whether the key was successfully deleted."}},"required":["prefix","deleted"]}},"required":["data"]}}}}}}}}} ``` # API Key Usage Query how much a given API key spent over a period of time.\ To make a request, you only need your AIMLAPI key obtained from your [account dashboard](https://aimlapi.com/app/keys). Both endpoints take the same parameters and differ only in what they return: `/v2/usage/detail` adds a per-model breakdown. ## Choosing the time window Pass **either** `period` **or** both `start` and `end` — not both at once: * `period` is relative — `24h`, `7d`, that is a positive integer followed by `h` or `d`. * `start` and `end` are absolute ISO-8601 timestamps. The window cannot exceed **92 days**. For a longer report, request several windows and add them up. {% hint style="info" %} A timestamp without a UTC offset is read as **UTC**, not as your local time. Pass an explicit offset (`2026-07-01T00:00:00+03:00`) if you mean something else. The response always echoes `start` and `end` resolved to UTC, so you can see exactly which window was measured. {% endhint %} ## Choosing the key | Your key | `key_prefix` | Result | | -------------- | ------------------------ | ---------------- | | any key | omitted | its own spend | | management key | provided | that key's spend | | regular key | provided, its own prefix | its own spend | | regular key | provided, another prefix | `403` | Any key can read its own spend. Reading the spend of a **different** key requires a management key. ## Get key usage Returns the total spend for a key over the requested window. ## GET /v2/usage > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v2/usage":{"get":{"operationId":"_v2_usage","parameters":[{"name":"period","in":"query","required":false,"description":"Relative window: a positive integer followed by `h` or `d`, for example `24h` or `7d`. Provide either `period` or both `start` and `end`, never both.","schema":{"type":"string"}},{"name":"start","in":"query","required":false,"description":"Window start, ISO-8601. A value without a UTC offset is read as UTC. Must be used together with `end`.","schema":{"type":"string"}},{"name":"end","in":"query","required":false,"description":"Window end, ISO-8601. Must not be earlier than `start`, and the window must not exceed 92 days.","schema":{"type":"string"}},{"name":"key_prefix","in":"query","required":false,"description":"Prefix of the key to report on. Omit it to get the spend of the key you are authenticating with. Reporting on a different key requires a management key.","schema":{"type":"string"}}],"responses":{"200":{"description":"Total spend for the key over the requested window.","content":{"application/json":{"schema":{"type":"object","properties":{"api_key_prefix":{"type":"string","nullable":true,"description":"The requested key prefix, or null when it was omitted and the answer is your own key's spend."},"start":{"type":"string","description":"Window start, resolved to UTC."},"end":{"type":"string","description":"Window end, resolved to UTC."},"spend":{"type":"number","description":"Total spend for the key over the window, in USD."},"currency":{"type":"string","description":"Spend currency (always USD)."},"requests":{"type":"integer","description":"Number of charged requests in the window."}},"required":["api_key_prefix","start","end","spend","currency","requests"]}}}}}}}}} ``` ## Get detailed key usage Returns the same total plus a per-model breakdown, sorted by spend. ## GET /v2/usage/detail > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v2/usage/detail":{"get":{"operationId":"_v2_usage_detail","parameters":[{"name":"period","in":"query","required":false,"description":"Relative window: a positive integer followed by `h` or `d`, for example `24h` or `7d`. Provide either `period` or both `start` and `end`, never both.","schema":{"type":"string"}},{"name":"start","in":"query","required":false,"description":"Window start, ISO-8601. A value without a UTC offset is read as UTC. Must be used together with `end`.","schema":{"type":"string"}},{"name":"end","in":"query","required":false,"description":"Window end, ISO-8601. Must not be earlier than `start`, and the window must not exceed 92 days.","schema":{"type":"string"}},{"name":"key_prefix","in":"query","required":false,"description":"Prefix of the key to report on. Omit it to get the spend of the key you are authenticating with. Reporting on a different key requires a management key.","schema":{"type":"string"}}],"responses":{"200":{"description":"Total spend for the key over the requested window, with a per-model breakdown.","content":{"application/json":{"schema":{"type":"object","properties":{"api_key_prefix":{"type":"string","nullable":true,"description":"The requested key prefix, or null when it was omitted and the answer is your own key's spend."},"start":{"type":"string","description":"Window start, resolved to UTC."},"end":{"type":"string","description":"Window end, resolved to UTC."},"spend":{"type":"number","description":"Total spend for the key over the window, in USD. This is the authoritative total: it is computed from the total, not by adding up models, so it can differ from that sum by a few nano-dollars."},"currency":{"type":"string","description":"Spend currency (always USD)."},"requests":{"type":"integer","description":"Number of charged requests in the window."},"models":{"type":"array","description":"Per-model breakdown, highest spend first.","items":{"type":"object","properties":{"model":{"type":"string","description":"Model name."},"spend":{"type":"number","description":"Spend for this model, in USD."},"requests":{"type":"integer","description":"Charged requests for this model."}},"required":["model","spend","requests"]}}},"required":["api_key_prefix","start","end","spend","currency","requests","models"]}}}}}}}}} ``` # Complete Model List ## Get Model List via API You can query the complete list of available models through this API.\ No API key is required for this request. You can also simply open [this list](https://api.aimlapi.com/models) in any web browser. ## GET /models > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/models":{"get":{"operationId":"ModelsController_getModels_v1","responses":{"200":{"description":"A list of available models.","content":{"application/json":{"schema":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"Unique identifier of the model."},"type":{"type":"string","description":"Model interaction type."},"info":{"type":"object","description":"Metadata describing the model.","properties":{"name":{"type":"string","description":"Human-readable model name."},"developer":{"type":"string","description":"Organization or company that developed the model."},"description":{"type":"string","description":"Short description of the model and its primary capabilities."},"contextLength":{"type":"integer","description":"Maximum supported context window size in tokens."},"maxTokens":{"type":"integer","description":"Maximum number of tokens that can be generated in a single response."},"url":{"type":"string","format":"uri","description":"Public model landing page on AIML API website."},"docs_url":{"type":"string","format":"uri","description":"Link to the official API documentation for this model."}},"required":["name","developer","description","url","docs_url"]},"features":{"type":"array","description":"List of supported features and API capabilities for the model.","items":{"type":"string"}},"endpoints":{"type":"array","description":"API endpoints through which this model can be accessed.","items":{"type":"string"}}},"required":["id","type","info","features","endpoints"]}}}}}}}}}} ``` ## Output Examples by Model Type As of early 2026, this endpoint returns a list of more than 400 models. Each item represents a single model identified by a unique ID. Depending on the model category (chat, video, etc.), the set of fields in each item may vary slightly, so below we provide representative examples from the main model categories. #### Example output item for a chat model Unlike other types of models, every chat model includes a non-empty `features` list that clearly shows what the model can do: support for streaming, instructions for SYSTEM or DEVELOPER roles besides the regular prompt, whether the model is described by the developer as “thinking”, etc. For more details on many of these, see the [CAPABILITIES](/capabilities/completion-or-chat-models) section of this documentation portal. {% code overflow="wrap" %} ```json { "id": "o3-mini", "type": "chat-completion", "info": { "name": "o3 mini", "developer": "Open AI", "description": "OpenAI o3-mini excels in reasoning tasks with advanced features like deliberative alignment and extensive context support.", "contextLength": 200000, "maxTokens": 100000, "url": "https://aimlapi.com/models/openai-o3-mini-api", "docs_url": "https://docs.aimlapi.com/api-references/text-models-llm/openai/o3-mini" }, "features": [ "openai/chat-completion", "openai/response-api", "openai/chat-assistant", "openai/chat-completion.function", "openai/chat-completion.message.refusal", "openai/chat-completion.message.system", "openai/chat-completion.message.developer", "openai/chat-completion.message.assistant", "openai/chat-completion.stream", "openai/chat-completion.max-completion-tokens", "openai/chat-completion.number-of-messages", "openai/chat-completion.stop", "openai/chat-completion.seed", "openai/chat-completion.reasoning", "openai/chat-completion.response-format" ], "endpoints": [ "/v1/chat/completions", "/v1/responses" ] } ``` {% endcode %} #### Example output item for an image model {% code overflow="wrap" %} ```json { "id": "flux/kontext-max/text-to-image", "type": "image", "info": { "name": "Flux Kontext Max", "developer": "Flux", "description": "A new Flux model optimized for maximum image quality.", "url": "https://aimlapi.com/models/flux-1-kontext-max", "docs_url": "https://docs.aimlapi.com/api-references/image-models/flux/flux-kontext-max-text-to-image" }, "features": [], "endpoints": [ "/v1/images/generations" ] } ``` {% endcode %} #### Example output item for a video model {% code overflow="wrap" %} ```json { "id": "veo2/image-to-video", "type": "video", "info": { "name": "Veo2 Image-to-Video", "description": "Veo2 Image-to-Video: Google's AI transforming still images into dynamic videos", "developer": "Google", "url": "https://aimlapi.com/models/veo-2-image-to-video-api", "docs_url": "https://docs.aimlapi.com/api-references/video-models/google/veo2-image-to-video" }, "features": [], "endpoints": [ "/v2/generate/video/google/generation", "/v2/video/generations" ] } ``` {% endcode %} # All Model IDs A full list of available models. {% hint style="info" %} If you need to select models based on specific parameters for your task, visit the [dedicated page on our official website](https://aimlapi.com/models/), which offers convenient filtering options. On the selected model’s page, you can find detailed technical and commercial information. {% endhint %} {% hint style="success" %} To fetch the complete model list via the API, see [the API reference](/api-references/service-endpoints/complete-model-list) for the relevant service endpoint. {% endhint %} The section **Full List of Model IDs** below lists the identifiers of all available and deprecated models, grouped by category. These IDs are used to specify the exact models in your code, like this:
If you already know the model ID, use the page search function (Ctrl+F for Win/Linux, Command+F for Mac) to locate it. The hyperlink will take you directly to the model's API Reference page. {% hint style="success" %} **New Model Request** Can't find the model you need? Join our [Discord community](https://discord.gg/8CwhkUuCR6) to propose new models for integration into our API offerings. Your contributions help us grow and serve you better. {% endhint %} ## Full List of Model IDs ### Text Models (LLM)
Model ID + API Reference linkDeveloperContextModel Card
gpt-3.5-turboOpen AI16,000Chat GPT 3.5 Turbo
gpt-3.5-turbo-0125Open AI16,000Chat GPT-3.5 Turbo 0125
gpt-3.5-turbo-1106Open AI16,000Chat GPT-3.5 Turbo 1106
gpt-4o-2024-08-06Open AI128,000GPT-4o-2024-08-06
gpt-4o-2024-05-13Open AI128,000GPT-4o-2024-05-13
gpt-4o-miniOpen AI128,000Chat GPT 4o mini
gpt-4o-mini-2024-07-18Open AI128,000GPT 4o mini
gpt-4o-search-previewOpen AI128,000GPT-4o Search Preview
gpt-4o-mini-search-previewOpen AI128,000GPT-4o Mini Search Preview
o1Open AI200,000OpenAI o1
openai/o3-2025-04-16Open AI200,000o3
o3-miniOpen AI200,000OpenAI o3 mini
openai/o3-proOpen AI200,000o3-pro
openai/gpt-4.1-2025-04-14Open AI1,000,000GPT-4.1
openai/gpt-4.1-mini-2025-04-14Open AI1,000,000GPT-4.1 Mini
openai/gpt-4.1-nano-2025-04-14Open AI1,000,000GPT-4.1 Nano
openai/o4-mini-2025-04-16Open AI200,000GPT-o4-mini-2025-04-16
openai/gpt-oss-20bOpen AI128,000GPT OSS 20B
tencent/hy3Tencent262,144-
inclusionai/ling-3.0-flashinclusionAI262,144-
inclusionai/ling-3.0-tinyinclusionAI262,144-
deepseek/deepseek-v4-pro-0813DeepSeek AI1,048,576-
thinkingmachines/inklingThinking Machines1,048,576-
meta/muse-glimmer-30bMeta131,072-
thinkingmachines/inkling-smallThinking Machines524,288-
openai/gpt-oss-120bOpen AI128,000GPT OSS 120B
openai/gpt-latestOpen AI400,000GPT Latest
openai/gpt-mini-latestOpen AI400,000GPT Mini Latest
openai/gpt-5.6-lunaOpen AI1,050,000-
openai/gpt-5.6-luna-proOpen AI1,050,000-
openai/gpt-5.6-terra-proOpen AI1,050,000-
openai/gpt-5.6-terraOpen AI1,050,000-
openai/gpt-5.6-sol-proOpen AI1,050,000-
openai/gpt-5.6-solOpen AI1,050,000-
openai/gpt-chat-latestOpen AI400,000GPT Chat Latest
openai/gpt-5-2025-08-07Open AI400,000GPT-5
openai/gpt-5-mini-2025-08-07Open AI400,000GPT-5 Mini
openai/gpt-5-nano-2025-08-07Open AI400,000GPT-5 Nano
openai/gpt-5-chat-latestOpen AI400,000GPT-5 Chat
openai/gpt-5-1Open AI128,000GPT-5.1
openai/gpt-5-1-chat-latestOpen AI128,000GPT-5.1 Chat Latest
openai/gpt-5-1-codexOpen AI400,000GPT-5.1 Codex
openai/gpt-5-1-codex-miniOpen AI400,000GPT-5.1 Codex Mini
openai/gpt-5-2Open AI400,000GPT-5.2
openai/gpt-5-2-chat-latestOpen AI400,000GPT-5.2 Chat Latest
openai/gpt-5-2-proOpen AI400,000GPT-5.2 Pro
openai/gpt-5-2-codexOpen AI400,000GPT-5.2 Codex
openai/gpt-5-3-codexOpen AI400,000GPT-5.3 Codex
openai/gpt-5-4Open AI1,000,000GPT-5.4
openai/gpt-5-4-proOpen AI1,000,000GPT-5.4 Pro
openai/gpt-5-5Open AI1,000,000GPT-5.5
openai/gpt-5-5-proOpen AI1,000,000GPT 5.5 Pro
anthropic/claude-opus-4.1
claude-opus-4-1
claude-opus-4-1-20250805		Anthropic200,000Claude Opus 4.1

claude-sonnet-4-5-20250929

anthropic/claude-sonnet-4.5

claude-sonnet-4-5

Anthropic200,000Claude 4.5 Sonnet

anthropic/claude-haiku-4.5
claude-haiku-4-5

claude-haiku-4-5-20251001

Anthropic200,000Claude 4.5 Haiku
anthropic/claude-opus-4-5
claude-opus-4-5
claude-opus-4-5-20251101		Anthropic200,000Claude 4.5 Opus
anthropic/claude-opus-4-6Anthropic200,000Claude 4.6 Opus
anthropic/claude-sonnet-4.6
anthropic/claude-sonnet-4-6-20260218		Anthropic200,000Claude Sonnet 4.6
anthropic/claude-opus-4-7
claude-opus-4-7		Anthropic1,000,000Claude Opus 4.7
anthropic/claude-opus-4-8
claude-opus-4-8		Anthropic1,000,000Claude Opus 4.8
anthropic/claude-sonnet-5Anthropic1,000,000-
Qwen/Qwen2.5-7B-Instruct-TurboAlibaba Cloud32,000Qwen 2.5 7B Instruct Turbo
qwen-plusAlibaba Cloud131,000Qwen Plus
qwen-turboAlibaba Cloud1,000,000Qwen Turbo
alibaba/qwen3-32bAlibaba Cloud131,000Qwen3-32B
alibaba/qwen3-coder-480b-a35b-instructAlibaba Cloud262,000Qwen3 Coder
mistral/labs-leanstral-1-5Mistral AI256,000-
baidu/ernie-5.0Baidu128,000-
alibaba/glm-5.2Zhipu AI1,000,000-
alibaba/qwen3-235b-a22b-thinking-2507Alibaba Cloud262,000Qwen3 235B A22B Thinking
alibaba/qwen3-next-80b-a3b-instructAlibaba Cloud262,000Qwen3-Next-80B-A3B Instruct
alibaba/qwen3-next-80b-a3b-thinkingAlibaba Cloud262,000Qwen3-Next-80B-A3B Thinking
alibaba/qwen3-max-previewAlibaba Cloud258,000Qwen3-Max Preview
alibaba/qwen3-max-instructAlibaba Cloud262,000Qwen3-Max Instruct
qwen3-omni-30b-a3b-captionerAlibaba Cloud65,000qwen3-omni-30b-a3b-captioner
alibaba/qwen3-vl-32b-instructAlibaba Cloud126,000Qwen3 VL 32B Instruct
alibaba/qwen3-vl-32b-thinkingAlibaba Cloud126,000Qwen3 VL 32B Thinking
alibaba/qwen3.5-plus-20260218Alibaba Cloud1,000,000Qwen3.5 Plus
alibaba/qwen3.5-omni-plusAlibaba Cloud256,000Qwen3.5 Omni Plus
alibaba/qwen3.5-omni-flashAlibaba Cloud256,000Qwen3.5 Omni Flash
alibaba/qwen3.5-flashAlibaba Cloud262,144Coming Soon
alibaba/qwen3.6-27bAlibaba Cloud262,144Qwen 3.6 27B
alibaba/qwen3.6-35b-a3bAlibaba Cloud262,144Qwen3.6-35B-A3B
alibaba/qwen3.6-flashAlibaba Cloud1,000,000Qwen3.5 Flash
alibaba/qwen3.6-plusAlibaba Cloud1,000,000Qwen3.6-Plus
alibaba/qwen3.6-max-previewAlibaba Cloud262,144Qwen3.6 Max Preview
alibaba/qwen3.7-maxAlibaba Cloud1,000,000Qwen3.7 Max
alibaba/qwen3.8-maxAlibaba Cloud1,000,000-
alibaba/qwen3.7-plusAlibaba Cloud1,000,000Qwen3.7 Plus
anthracite-org/magnum-v4-72bAnthracite32,000Magnum v4 72B
baidu/ernie-4.5-vl-424b-a47bBaidu123,000ERNIE 4.5 VL
bytedance/seed-1-8ByteDance256,000Seed 1.8
bytedance/dola-seed-2-0-miniByteDance256,000Dola Seed 2.0 Mini
bytedance/dola-seed-2-0-liteByteDance256,000Dola Seed 2.0 Lite
bytedance/dola-seed-2-0-proByteDance256,000Dola Seed 2.0 Pro
bytedance/dola-seed-2-0-codeByteDance256,000Dola Seed 2.0 Code
cohere/command-aCohere256,000Command A
deepseek-chat or
deepseek/deepseek-chat or
deepseek/deepseek-chat-v3-0324		DeepSeek128,000DeepSeek V3
deepseek/deepseek-r1 or
deepseek-reasoner		DeepSeek128,000DeepSeek R1
deepseek/deepseek-chat-v3.1DeepSeek128,000DeepSeek V3.1 Chat
deepseek/deepseek-reasoner-v3.1DeepSeek128,000DeepSeek V3.1 Reasoner
deepseek/deepseek-thinking-v3.2-expDeepSeek128,000DeepSeek V3.2-Exp Thinking
deepseek/deepseek-non-thinking-v3.2-expDeepSeek128,000DeepSeek V3.2-Exp Non-Thinking
deepseek/deepseek-reasoner-v3.1-terminusDeepSeek128,000DeepSeek V3.1 Terminus Reasoning
deepseek/deepseek-non-reasoner-v3.1-terminusDeepSeek128,000DeepSeek V3.1 Terminus Non-Reasoning
deepseek/deepseek-v4-proDeepSeek1,000,000DeepSeek V4 Pro
deepseek/deepseek-v4-flashDeepSeek1,000,000DeepSeek 4 Flash
google/gemini-2.5-flash-lite-previewGoogle1,000,000–
google/gemini-2.5-flashGoogle1,000,000Gemini 2.5 Flash
google/gemini-3-flash-previewGoogle1,000,000Gemini 3 Flash
google/gemini-2.5-proGoogle1,000,000Gemini 2.5 Pro
google/gemma-4-26b-a4b-it-maasGoogle262,144-
google/gemini-3-6-flashGoogle1,048,576-
google/gemini-3-7-flashGoogle1,048,576-
google/gemini-3-5-flash-liteGoogle1,048,576-
google/gemini-3.5-flash-liteGoogle1,048,576-
google/gemma-3-4b-itGoogle128,000Gemma 3 (4B)
google/gemma-3-12b-itGoogle128,000Gemma 3 (12B)
google/gemma-3-27b-itGoogle128,000Gemma 3 (27B)
google/gemma-3n-e4b-itGoogle8,192Gemma 3n 4B
google/gemini-3-1-pro-previewGoogle1,000,000Gemini 3.1 Pro
google/gemini-3-1-flash-lite-previewGoogle1,048,576Gemini 3.1 Flash Lite
google/gemini-3-1-flash-liteGoogle1,050,000Gemini 3.1 Flash Lite
google/gemma-4-31b-itGoogle262,000Gemma 4 31B
google/gemini-3-5-flashGoogle1,000,000Gemini 3.5 Flash
google/gemini-3-pro-imageGoogle65,536-
google/gemma-2-27b-itGoogle8,192-
google/gemini-3.1-pro-preview-customtoolsGoogle1,048,756-
google/gemini-3.1-flash-imageGoogle131,072-
gryphe/mythomax-l2-13bGryphe4,000MythoMax-L2 (13B)
meta-llama/Llama-3.3-70B-Instruct-TurboMeta128,000Meta Llama 3.3 70B Instruct Turbo
meta-llama/llama-3.3-70b-versatileMeta131,000Llama 3.3 70B Versatile
MiniMax-Text-01MiniMax1,000,000MiniMax-Text-01
minimax/m1MiniMax1,000,000MiniMax M1
minimax/m2MiniMax200,000MiniMax M2
minimax/m2-herMiniMax200,000MiniMax M2-her
minimax/m2-1MiniMax204,800MiniMax M2.1
minimax/m2-1-highspeedMiniMax204,800MiniMax M2.1 Highspeed
minimax/m2-5-20260218MiniMax204,800MiniMax M2.5
minimax/m2-5-highspeed-20260218MiniMax204,800MiniMax M2.5
minimax/m2-7-20260402MiniMax204,800MiniMax M2.7
minimax/m2-7-highspeedMiniMax204,800MiniMax M2.7 Highspeed
minimax/minimax-m3MiniMax1,000,000MiniMax M3
mistralai/mistral-nemoMistral AI128,000Mistral Nemo
moonshot/kimi-k2-5Moonshot262,000Kimi K2.5
moonshot/kimi-k2-6Moonshot256,000Kimi K2.6
moonshot/kimi-k2-7-codeMoonshot262,000Kimi K2.7 Code
moonshot/kimi-k2-7-code-highspeedMoonshot262,144Kimi K2.7 Code Highspeed
meituan/longcat-2.0Meituan1,000,000-
sakana/fugu-ultraSakana AI1,000,000Sakana Fugu Ultra
sakana/sakana-namazuSakana AI262,144-
alibaba/qwen3.8-2.4t-a95bAlibaba Cloud262,144-
nousresearch/hermes-4-405bNousResearch131,000Hermes 4 405b
nvidia/nemotron-3-ultra-550b-a55bNVIDIA1,000,000Nemotron 3 Ultra 550b
nvidia/nemotron-3-nano-30b-a3bNVIDIA1,000,000Nemotron 3 Nano 30B
nvidia/nemotron-3.5-lightningNvidia1,000,000-
nvidia/nemotron-3-nano-omni-30b-a3b-reasoningNVIDIA1,000,000Nemotron 3 Nano Omni
nvidia/nemotron-3-super-120b-a12bNVIDIA1,000,000Nemotron 3 Super 120B
perplexity/sonarPerplexity128,000Sonar
perplexity/sonar-proPerplexity200,000Sonar Pro
stepfun/step-3.7-flashStepFun256,000Step 3.7 Flash
x-ai/grok-3-betaxAI131,000Grok 3 Beta
x-ai/grok-3-mini-betaxAI131,000Grok 3 Beta Mini
x-ai/grok-code-fast-1xAI256,000Grok Code Fast 1
x-ai/grok-4-fast-non-reasoningxAI2,000,000Grok 4 Fast
x-ai/grok-4-fast-reasoningxAI2,000,000Grok 4 Fast Reasoning
x-ai/grok-4-1-fast-non-reasoningxAI2,000,000Grok 4.1 Fast Non-Reasoning
x-ai/grok-4-1-fast-reasoningxAI2,000,000Grok 4.1 Fast Reasoning
x-ai/grok-4-5X AI500,000-
x-ai/grok-4-6X AI500,000-
x-ai/grok-4-20-0309-non-reasoningxAI2,000,000Grok 4.20 Non-Reasoning
x-ai/grok-4-20-0309-reasoningxAI2,000,000Grok 4.20 Reasoning
x-ai/grok-4-3X AI1,000,000-
x-ai/grok-build-0-1xAI256,000Grok Build 0.1
xiaomi/mimo-v2.5Xiaomi1,000,000MiMo V2.5
ai21/jamba-large-1.7AI21256,000-
allenai/olmo-3-32b-thinkAllenAI65,536-
amazon/nova-2-lite-v1Amazon1,000,000-
deepseek/deepseek-v4-flash-latestDeepSeek AI1,048,576-
bytedance-seed/seed-2-1-turboByteDance262,144-
bytedance-seed/seed-2.0-codeByteDance262,144-
meta/muse-spark-1.1Meta1,048,576-
meta/muse-spark-1.2Meta1,048,576-
cognitivecomputations/dolphin-mistral-24b-venice-editionVenice128,000-
anthropic/claude-3-haikuAnthropic200,000-
anthropic/claude-sonnet-latestAnthropic1,000,000-
anthropic/claude-opus-latestAnthropic1,000,000-
anthropic/claude-opus-5Anthropic1,000,000-
anthropic/claude-haiku-latestAnthropic1,000,000-
anthropic/claude-fable-latestAnthropic1,000,000-
anthropic/claude-opus-4.8-fastAnthropic1,000,000-
anthropic/claude-sonnet-4Anthropic1,000,000-
z-ai/glm-4.5vZhipu AI65,536-
z-ai/glm-5v-turboZhipu AI202,752-
z-ai/glm-5-turboZhipu AI262,144-
z-ai/glm-4.7-flashZhipu AI202,752-
z-ai/glm-4.6vZhipu AI131,072-
writer/palmyra-x5Writer1,040,000-
upstage/solar-pro-3Upstage128,000-
upstage/solar-pro4Upstage524,288-
undi95/remm-slerp-l2-13bundi956,144-
thedrummer/cydonia-24b-v4.1TheDrummer131,072-
thedrummer/rocinante-12bTheDrummer32,768-
thedrummer/unslopnemo-12bTheDrummer32,768-
thedrummer/skyfall-36b-v2TheDrummer32,768-
switchpoint/routerswitchpoint131,072-
openai/gpt-4-turbo-previewOpen AI128,000-
openai/gpt-5.2-chatOpen AI128,000-
openai/gpt-5-imageOpen AI400,000-
openai/gpt-5.4-nanoOpen AI400,000-
openai/gpt-5.4-proOpen AI1,050,000-
nousresearch/hermes-3-llama-3.1-405bNousResearch131,072-
nousresearch/hermes-3-llama-3.1-70bNousResearch131,072-
nousresearch/hermes-4-70bNousResearch131,072-
nex-agi/nex-n2-miniNex AGI262,144-
nex-agi/nex-n2-proNex AGI262,144-
morph/morph-v3-fastMorph81,920-
morph/morph-v3-largeMorph262,144-
moonshotai/kimi-k2Moonshot131,072-
moonshotai/kimi-k2-thinkingMoonshot262,144-
moonshotai/kimi-k2-0905Moonshot262,144-
mistralai/codestral-2508Mistral AI256,000-
mistralai/mistral-sabaMistral AI32,768-
mistralai/voxtral-small-24b-2507Mistral AI32,000-
mistralai/mixtral-8x22b-instructMistral AI65,536-
mistralai/mistral-small-3.2-24b-instructMistral AI128,000-
mistralai/mistral-small-3.1-24b-instructMistral AI128,000-
mistralai/mistral-small-2603Mistral AI262,144-
mistralai/mistral-small-24b-instruct-2501Mistral AI32,768-
mistralai/mistral-medium-3.1Mistral AI131,072-
mistralai/ministral-8b-2512Mistral AI262,144-
mistralai/ministral-3b-2512Mistral AI131,072-
mistralai/ministral-14b-2512Mistral AI262,144-
mistralai/devstral-2512Mistral AI262,144-
mancer/weaverMancer8,000-
liquid/lfm-2-24b-a2bLiquidAI128,000-
kwaipilot/kat-coder-pro-v2Kwaipilot256,000-
kwaipilot/kat-coder-air-v2.5Kwaipilot256,000-
kwaipilot/kat-coder-pro-v2.5Kwaipilot256,000-
inclusionai/ling-2.6-1tinclusionAI262,144-
inclusionai/ring-2.6-1tinclusionAI262,144-
inclusionai/ling-2.6-flashinclusionAI262,144-
inception/mercury-2Inception128,000-
ibm-granite/granite-4.0-h-microIBM131,000-
ibm-granite/granite-4.1-8bIBM131,072-
cohere/command-r-08-2024Cohere128,000-
bytedance-seed/seed-1.6ByteDance Seed262,144-
bytedance-seed/seed-2.0-liteByteDance Seed262,144-
bytedance-seed/seed-2.0-miniByteDance Seed262,144-
bytedance-seed/seed-1.6-flashByteDance Seed262,144-
arcee-ai/trinity-large-thinkingArcee AI262,144-
arcee-ai/virtuoso-largeArcee AI131,072-
arcee-ai/trinity-miniArcee AI131,072-
amazon/nova-pro-v1Amazon300,000-
amazon/nova-premier-v1Amazon1,000,000-
amazon/nova-micro-v1Amazon128,000-
amazon/nova-lite-v1Amazon300,000-
aion-labs/aion-rp-llama-3.1-8bAionLabs32,768-
aion-labs/aion-2.0AionLabs131,072-
aion-labs/aion-3.0-miniAionLabs131,072-
xiaomi/mimo-v2.5-proXiaomi1,000,000MiMo V2.5 Pro
zhipu/glm-4.5-airZhipu128,000GLM-4.5 Air
zhipu/glm-4.5Zhipu128,000GLM-4.5
zhipu/glm-4.6Zhipu200,000GLM-4.6
zhipu/glm-4.7Zhipu200,000GLM-4.7
zhipu/glm-5Zhipu200,000GLM-5
zhipu/glm-5-1Zhipu200,000GLM-5.1
zhipu/glm-5-2Zhipu200,000GLM-5.2
poolside/laguna-s-2.1Poolside1,048,576-
## Image Models
Model ID + API Reference linkDeveloperContextModel Card
alibaba/qwen-imageAlibaba CloudQwen Image
alibaba/qwen-image-editAlibaba CloudQwen Image Edit
alibaba/z-image-turboAlibaba CloudZ-Image Turbo
alibaba/z-image-turbo-loraAlibaba CloudZ-Image Turbo LoRA
alibaba/wan2.2-t2i-plusAlibaba CloudWan 2.2 Plus
alibaba/wan2.2-t2i-flashAlibaba CloudWan 2.2 Flash
alibaba/wan2.5-t2i-previewAlibaba CloudWan 2.5 Preview
alibaba/wan-2-6-imageAlibaba CloudWan 2.6
alibaba/wan-2-7-imageAlibaba CloudWan 2.7 Image
alibaba/wan-2-7-image-proAlibaba CloudWan 2.7 Image Pro
alibaba/qwen-image-2-0Alibaba CloudQwen Image 2.0
alibaba/qwen-image-2-0-proAlibaba CloudQwen Image 2.0 Pro
bytedance/dreamina-creative-upscaleByteDanceDreamina Creative Upscale
bytedance/seedream-v4-text-to-imageByteDanceSeedream 4 Text-to-Image
bytedance/seedream-v4-editByteDanceSeedream 4 Edit
bytedance/usoByteDanceUSO
bytedance/seedream-4-5ByteDanceSeedream 4.5
bytedance/seedream-5-0-lite-previewByteDanceSeadream 5.0 Lite
bytedance/seedream-5-0-proByteDance-
bytedance/dola-seedream-5-0-proByteDance-
flux-proFluxFLUX.1 [pro]
flux-pro/v1.1FluxFLUX 1.1 [pro]
flux-pro/v1.1-ultraFluxFLUX 1.1 [pro ultra]
flux-realismFluxFLUX Realism LoRA
flux/devFluxFLUX.1 [dev]
flux/dev/image-to-imageFlux-
flux/schnellFluxFLUX.1 [schnell]
flux/kontext-max/text-to-imageFluxFLUX.1 Kontext [max]
flux/kontext-max/image-to-imageFluxFLUX.1 Kontext [max]
flux/kontext-pro/text-to-imageFluxFlux.1 Kontext [pro]
flux/kontext-pro/image-to-imageFluxFlux.1 Kontext [pro]
beeble/switchx-image-to-imageBeeble-
flux/srpoFluxFLUX.1 SRPO Text-to-Image
flux/srpo/image-to-imageFluxFLUX.1 SRPO Image-to-Image
blackforestlabs/flux-2FluxFLUX.2
blackforestlabs/flux-2-editFluxFLUX.2 Edit
blackforestlabs/flux-2-loraFluxFlux 2 LoRA
blackforestlabs/flux-2-lora-editFluxFlux 2 LoRA Edit
blackforestlabs/flux-2-proFluxFLUX.2 [pro]
blackforestlabs/flux-2-pro-editFluxFLUX.2 [pro] Edit
imagen-3.0-generate-002GoogleImagen 3
google/nano-banana-editGoogle-
google/imagen4/previewGoogleImagen 4 Preview
google/imagen-4.0-generate-001GoogleImagen 4.0 Generate
google/imagen-4.0-fast-generate-001GoogleImagen 4.0 Fast Generate
google/imagen-4.0-ultra-generate-001GoogleImagen 4.0 Ultra Generate
google/gemini-2.5-flash-imageGoogleGemini 2.5 Flash Image
imagen4/previewGoogle-
google/gemini-2.5-flash-image-editGoogleGemini 2.5 Flash Image Edit
google/nano-banana-pro
google/gemini-3-pro-image-preview		GoogleGemini 3 Pro Image (Nano Banana Pro)
google/nano-banana-pro-edit
google/gemini-3-pro-image-preview-edit		GoogleGemini 3 Pro Image Edit (Nano Banana Pro)
google/gemini-3-pro-image-editGoogle-
google/nano-banana-2
google/gemini-3-1-flash-image-preview		GoogleGemini 3.1 Flash Image (Nano Banana 2)
google/gemini-3-1-flash-lite-imageGoogle-
klingai/image-o1Kling AIKling Image O1
luma/uni-1Luma AI-
luma/uni-1-maxLuma AI-
microsoft/mai-image-2.5MicrosoftMAI-Image 2.5
openai/gpt-image-1OpenAIgpt-image-1
openai/gpt-image-1-miniOpenAIGPT Image 1 Mini
openai/gpt-image-1-5OpenAIGPT Image 1.5
openai/gpt-image-2OpenAIGPT Image 2
recraft-v3Recraft AIRecraft v3
reve/create-imageReveReve Create Image
reve/edit-imageReveReve Edit Image
reve/remix-edit-imageReveReve Remix Image
stable-diffusion-v3-mediumStability AIStable Diffusion 3
stable-diffusion-v35-largeStability AIStable Diffusion 3.5 Large
hunyuan/hunyuan-image-v3-text-to-imageTencentHunyuanImage 3.0
alibaba/qwen-image-3Alibaba Cloud-
alibaba/qwen-image-3-editAlibaba Cloud-
topaz-labs/sharpenTopaz LabsSharpen
topaz-labs/sharpen-genTopaz LabsSharpen Generative
x-ai/grok-imagine-imagexAIGrok Imagine
x-ai/grok-imagine-image-proxAIGrok Imagine Image Pro
### Video Models
Model ID + API Reference linkDeveloperContextModel Card
alibaba/wan2.1-t2v-plusAlibaba CloudWan2.1 Plus
alibaba/wan2.1-t2v-turboAlibaba CloudWan2.1 Turbo
alibaba/wan2.2-t2v-plusAlibaba CloudWan 2.2 T2V
alibaba/wan2.5-t2v-previewAlibaba CloudWan 2.5 Text-to-Video
alibaba/wan2.5-i2v-previewAlibaba CloudWan 2.5 Image-to-Video
alibaba/wan2.2-14b-animate-replaceAlibaba CloudWan 2.2 14b animate replace
alibaba/wan2.2-14b-animate-moveAlibaba CloudWan 2.2 14b animate move
alibaba/wan2.2-vace-fun-a14b-reframeAlibaba CloudWan 2.2 vace fun 14b reframe
alibaba/wan2.2-vace-fun-a14b-outpaintingAlibaba CloudWan 2.2 vace fun 14b outpainting
alibaba/wan2.2-vace-fun-a14b-inpaintingAlibaba CloudWan 2.2 vace fun 14b inpainting
alibaba/wan2.2-vace-fun-a14b-poseAlibaba CloudWan 2.2 vace fun 14b pose
alibaba/wan2.2-vace-fun-14b-depthAlibaba CloudWan 2.2 vace fun 14b depth
alibaba/wan2.5-t2v-previewAlibaba CloudWan 2.5 Preview
alibaba/wan2.5-i2v-previewAlibaba Cloud-
alibaba/wan-2-6-t2vAlibaba CloudWan 2.6 Text-to-Video
alibaba/wan-2-6-i2vAlibaba CloudWan 2.6 Image-to-Video
alibaba/wan-2-6-r2vAlibaba CloudWan 2.6 Reference-to-Video
alibaba/wan-2-6-image-to-video-flashAlibaba CloudWan 2.6 Image-to-Video Flash
alibaba/happyhorse-1-0Alibaba CloudHappy Horse 1.0
alibaba/happyhorse-1.1Alibaba CloudHappy Horse 1.1
alibaba/wan-2-7-t2vAlibaba CloudWan 2.7
alibaba/wan-2-7-i2vAlibaba CloudWan 2.7
alibaba/wan-2-7-r2vAlibaba CloudWan 2.7
bytedance/seedance-1-0-pro-t2vByteDanceSeedance 1.0 Pro
bytedance/seedance-1-0-pro-i2vByteDanceSeedance 1.0 Pro
bytedance/seedance-1-0-pro-fastByteDanceSeedance 1.0 Pro Fast
bytedance/omnihumanByteDanceOmniHuman
bytedance/omnihuman/v1.5ByteDanceOmniHuman v1.5
bytedance/seedance-1-5-proByteDanceSeedance 1.5 Pro
bytedance/seedance-2-0ByteDanceSeedance 2.0
bytedance/seedance-2-5ByteDance-
bytedance/dreamina-seedance-2-5ByteDance-
google/gemini-omni-flash-previewGoogle-
bytedance/dreamina-seedance-2-0-miniByteDanceSeedance 2.0 Mini
bytedance/seedance-2-0-fastByteDanceSeedance 2.0 Fast
veo2GoogleVeo2 Text-to-Video
veo2/image-to-videoGoogleVeo2 Image-to-Video
google/veo3GoogleVeo 3
google/veo-3.0-i2vGoogleVeo 3 I2V
google/veo-3.0-fastGoogleVeo 3 Fast
google/veo-3.0-i2v-fastGoogleVeo 3 I2V Fast
google/veo-3.1-t2vGoogleVeo 3.1 Text-to-Video
google/veo-3.1-t2v-fastGoogleVeo 3.1 Fast Text-to-Video
google/veo-3.1-i2vGoogleVeo 3.1 Image-to-Video
google/veo-3.1-i2v-fastGoogleVeo 3.1 Fast Image-to-Video
google/veo-3.1-reference-to-videoGoogleVeo 3.1 Reference-to-Video
google/veo-3.1-first-last-image-to-videoGoogleVeo 3.1 First-Last Frame-to-Video
google/veo-3.1-first-last-image-to-video-fastGoogleVeo 3.1 Fast First-Last Frame-to-Video
google/veo3-1-extend-videoGoogleVeo 3.1 Extend Video
google/veo3-1-fast-extend-videoGoogleVeo 3.1 Fast Extend Video
google/veo-3-1-lite-generate-previewGoogleVeo 3.1 Lite
google/veo-3.1-lite-generate-001GoogleVeo 3.1 Lite
kling-video/v1/standard/image-to-videoKling AIKling AI (image-to-video)
kling-video/v1/standard/text-to-videoKling AIKling AI (text-to-video)
kling-video/v1/pro/image-to-videoKling AIKling AI (image-to-video)
kling-video/v1/pro/text-to-videoKling AIKling AI (text-to-video)
kling-video/v1.6/standard/text-to-videoKling AIKling 1.6 Standard
kling-video/v1.6/standard/image-to-videoKling AIKling 1.6 Standard
kling-video/v1.6/pro/image-to-videoKling AIKling 1.6 Pro
kling-video/v1.6/pro/text-to-videoKling AIKling 1.6 Pro
klingai/kling-video-v1.6-pro-effectsKling AIKling 1.6 Pro Effects
klingai/kling-video-v1.6-standard-effectsKling AIKling 1.6 Standard Effects
kling-video/v1.6/standard/multi-image-to-videoKling AIKling V1.6 Multi-Image-to-Video
klingai/v2-master-image-to-videoKling AIKling 2.0 Master
klingai/v2-master-text-to-videoKling AIKling 2.0 Master
kling-video/v2.1/standard/image-to-videoKling AIKling V2.1 Standard I2V
kling-video/v2.1/pro/image-to-videoKling AIKling V2.1 Pro I2V
klingai/v2.1-master-image-to-videoKling AIling 2.1 Master
klingai/v2.1-master-text-to-videoKling AIKling 2.1 Master
klingai/v2.5-turbo/pro/image-to-videoKling AIKling Video v2.5 Turbo Pro Image-to-Video
klingai/v2.5-turbo/pro/text-to-videoKling AIKling Video v2.5 Turbo Pro Text-to-Video
klingai/avatar-standardKling AIKling AI Avatar Standard
klingai/avatar-proKling AIKling AI Avatar Pro
klingai/video-v2-6-pro-text-to-videoKling AIKling 2.6 Pro Text-to-Video
klingai/video-v2-6-pro-image-to-videoKling AIKling 2.6 Pro Image-to-Video
klingai/video-o1-image-to-videoKling AIKling Video O1 Image to Video
klingai/video-o1-reference-to-videoKling AIKling Video O1 Reference-to-Video
klingai/video-o1-video-to-video-editKling AIKling Video O1 Video to Video Edit
klingai/video-o1-video-to-video-referenceKling AIKling Video O1 Video-to-Video Reference
klingai/video-v2-6-pro-motion-controlKling AIKling 2.6 Pro Motion Control
klingai/video-v3-standard-text-to-videoKling AIKling Video v3 Standard
klingai/video-v3-standard-image-to-videoKling AIKling Video v3 Standard
klingai/video-v3-omni-720p-image-to-videoKling AI-
klingai/video-v3-omni-1080p-image-to-videoKling AI-
klingai/video-v3-pro-text-to-videoKling AIKling Video v3 Pro
klingai/video-v3-omni-720p-text-to-videoKling AI-
klingai/video-v3-omni-1080p-text-to-videoKling AI-
klingai/video-v3-pro-image-to-videoKling AIKling Video v3 Pro
klingai/video-v3-turbo-pro-image-to-videoKling AIKling Video V3 Standard Turbo Image to Video
klingai/video-v3-turbo-pro-text-to-videoKling AIKling Video V3 Turbo Pro Text to Video
klingai/video-v3-standard-turbo-text-to-videoKling AIKling Video V3 Standard Turbo Text to Video
klingai/video-v3-turbo-pro-image-to-videoKling AIKling Video V3 Turbo Pro Image to Video
krea/krea-wan-14b/text-to-videoKreaKrea WAN 14B Text-to-Video
krea/krea-wan-14b/video-to-videoKreaKrea WAN 14B Video-to-Video
beeble/switchx-video-to-videoBeeble-
ltxv/ltxv-2LTXVLTXV 2
ltxv/ltxv-2-fastLTXVLTXV 2 Fast
luma/ray-2Luma AIRay 2
luma/ray-flash-2Luma AIRay Flash 2
luma/ray-3.2Luma AI-
magic/text-to-videoMagicMagic Video
magic/image-to-videoMagicMagic Video
magic/video-to-videoMagicMagic Video
video-01MiniMaxMiniMax Video-01
video-01-live2dMiniMax-
minimax/hailuo-02MiniMaxHailuo 02
minimax/h3Minimax AI-
minimax/hailuo-2.3MiniMaxHailuo 2.3
minimax/hailuo-2.3-fastMiniMaxHailuo 2.3 Fast
sora-2-t2vOpenAI-
sora-2-i2vOpenAI-
sora-2-pro-t2vOpenAI-
sora-2-pro-i2vOpenAI-
pixverse/v5/text-to-videoPixVersePixverse v5 Text-to-Video
pixverse/v5/image-to-videoPixVersePixverse v5 Image-to-Video
pixverse/v5/transitionPixVersePixverse v5 Transition
pixverse/v5-5-text-to-videoPixVersePixVerse V5.5 Text-to-Video
pixverse/v5-5-image-to-videoPixVersePixverse v5.5 Image-to-Video
pixverse/lip-syncPixVerse-
gen3a_turboRunwayRunway Gen-3 turbo
runway/gen4_turboRunwayRunway Gen-4 Turbo
runway/gen4_alephRunwayAleph
runway/act_twoRunwayRunway Act Two
sber-ai/kandinsky5-t2vSber AIKandinsky 5 Standard
sber-ai/kandinsky5-distill-t2vSber AIKandinsky 5 Distill
tencent/hunyuan-video-foleyTencentHunyuanVideo Foley
veed/fabric-1.0Veedfabric-1.0
veed/fabric-1.0-fastVeedfabric-1.0-fast
x-ai/grok-imagine-videoxAIGrok Imagine Video
x-ai/grok-imagine-video-1.5-previewxAIGrok Imagine Video 1.5
blackforestlabs/flux-3-video-draft-i2vBlack Forest Labs-
blackforestlabs/flux-3-video-i2vBlack Forest Labs-
blackforestlabs/flux-3-video-draft-t2vBlack Forest Labs-
blackforestlabs/flux-3-video-t2vBlack Forest Labs-
blackforestlabs/flux-3-video-draft-v2vBlack Forest Labs-
blackforestlabs/flux-3-video-v2vBlack Forest Labs-
### Voice/Speech Models #### Speech-to-Text
Model ID + API Reference linkDeveloperContextModel Card
aai/slam-1Assembly AISlam 1
aai/universalAssembly AIUniversal
nova-3DeepgramNova-3
nova-3-generalDeepgramNova-3 General
nova-3-medicalDeepgramNova-3 Medical
#g1_nova-2-automotiveDeepgramDeepgram Nova-2
#g1_nova-2-conversationalaiDeepgramDeepgram Nova-2
#g1_nova-2-drivethruDeepgramDeepgram Nova-2
#g1_nova-2-financeDeepgramDeepgram Nova-2
#g1_nova-2-generalDeepgramDeepgram Nova-2
#g1_nova-2-medicalDeepgramDeepgram Nova-2
#g1_nova-2-meetingDeepgramDeepgram Nova-2
#g1_nova-2-phonecallDeepgramDeepgram Nova-2
#g1_nova-2-videoDeepgramDeepgram Nova-2
#g1_nova-2-voicemailDeepgramDeepgram Nova-2
#g1_whisper-tinyOpenAI-
#g1_whisper-smallOpenAI-
#g1_whisper-baseOpenAI-
#g1_whisper-mediumOpenAI-
#g1_whisper-largeOpenAIWhisper
openai/gpt-4o-transcribeOpenAIGPT-4o Transcribe
openai/gpt-4o-mini-transcribeOpenAIGPT-4o Mini Transcribe
microsoft/mai-transcribe-1.5MicrosoftMAI-Transcribe-1.5
#### Text-to-Speech
Model IDDeveloperContextModel Card
alibaba/qwen3-tts-flashAlibaba CloudQwen3-TTS-Flash
#g1_aura-angus-enDeepgramAura
#g1_aura-arcas-enDeepgramAura
#g1_aura-asteria-enDeepgramAura
#g1_aura-athena-enDeepgramAura
#g1_aura-helios-enDeepgramAura
#g1_aura-hera-enDeepgramAura
#g1_aura-luna-enDeepgramAura
#g1_aura-orion-enDeepgramAura
#g1_aura-orpheus-enDeepgramAura
#g1_aura-perseus-enDeepgramAura
#g1_aura-stella-enDeepgramAura
#g1_aura-zeus-enDeepgramAura
#g1_aura-2-amalthea-enDeepgramAura 2
#g1_aura-2-andromeda-enDeepgramAura 2
#g1_aura-2-apollo-enDeepgramAura 2
#g1_aura-2-arcas-enDeepgramAura 2
#g1_aura-2-aries-enDeepgramAura 2
#g1_aura-2-asteria-enDeepgramAura 2
#g1_aura-2-athena-enDeepgramAura 2
#g1_aura-2-atlas-enDeepgramAura 2
#g1_aura-2-aurora-enDeepgramAura 2
#g1_aura-2-callista-enDeepgramAura 2
#g1_aura-2-cora-enDeepgramAura 2
#g1_aura-2-cordelia-enDeepgramAura 2
#g1_aura-2-delia-enDeepgramAura 2
#g1_aura-2-draco-enDeepgramAura 2
#g1_aura-2-electra-enDeepgramAura 2
#g1_aura-2-harmonia-enDeepgramAura 2
#g1_aura-2-helena-enDeepgramAura 2
#g1_aura-2-hera-enDeepgramAura 2
#g1_aura-2-hermes-enDeepgramAura 2
#g1_aura-2-hyperion-enDeepgramAura 2
#g1_aura-2-iris-enDeepgramAura 2
#g1_aura-2-janus-enDeepgramAura 2
#g1_aura-2-juno-enDeepgramAura 2
#g1_aura-2-jupiter-enDeepgramAura 2
#g1_aura-2-luna-enDeepgramAura 2
#g1_aura-2-mars-enDeepgramAura 2
#g1_aura-2-minerva-enDeepgramAura 2
#g1_aura-2-neptune-enDeepgramAura 2
#g1_aura-2-odysseus-enDeepgramAura 2
#g1_aura-2-ophelia-enDeepgramAura 2
#g1_aura-2-orion-enDeepgramAura 2
#g1_aura-2-orpheus-enDeepgramAura 2
#g1_aura-2-pandora-enDeepgramAura 2
#g1_aura-2-phoebe-enDeepgramAura 2
#g1_aura-2-pluto-enDeepgramAura 2
#g1_aura-2-saturn-enDeepgramAura 2
#g1_aura-2-selene-enDeepgramAura 2
#g1_aura-2-thalia-enDeepgramAura 2
#g1_aura-2-theia-enDeepgramAura 2
#g1_aura-2-vesta-enDeepgramAura 2
#g1_aura-2-zeus-enDeepgramAura 2
#g1_aura-2-celeste-esDeepgramAura 2
#g1_aura-2-estrella-esDeepgramAura 2
#g1_aura-2-nestor-esDeepgramAura 2
elevenlabs/eleven_multilingual_v2ElevenLabsElevenLabs Multilingual v2
elevenlabs/eleven_turbo_v2_5ElevenLabsElevenLabs Turbo v2.5
hume/octave-2Hume AIOctave 2
inworld/tts-1InworldInworld TTS-1
inworld/tts-1-maxInworldInworld TTS-1-Max
inworld/tts-1-5-miniInworldInworld TTS-1.5-Mini
inworld/tts-1-5-maxInworldComing Soon
microsoft/vibevoice-1.5bMicrosoftVibeVoice 1.5B
microsoft/vibevoice-7bMicrosoftVibeVoice 7B
microsoft/mai-voice-2MicrosoftMAI Voice-2
bytedance/seed-audio-1-0ByteDance-
openai/tts-1OpenAITTS-1
openai/tts-1-hdOpenAITTS-1 HD
openai/gpt-4o-mini-ttsOpenAIGPT-4o-mini-TTS
#### Voice Chat
Model IDDeveloperContextModel Card
elevenlabs/v3_alphaElevenLabsEleven v3 Alpha
minimax/speech-2.5-turbo-previewMiniMaxMiniMax Speech 2.5 Turbo
minimax/speech-2.5-hd-previewMiniMaxMiniMax Speech 2.5 HD
minimax/speech-2.6-turboMiniMaxMiniMax Speech 2.6 Turbo
minimax/speech-2.6-hdMiniMaxMiniMax Speech 2.6 HD
minimax/speech-2.8-turboMiniMaxSpeech 2.8 Turbo
minimax/speech-2.8-hdMiniMaxSpeech 2.8 HD
### Music Models
Model IDDeveloperContextModel Card
elevenlabs/eleven_musicElevenLabsEleven Music
google/lyria2GoogleLyria 2
stable-audioStability AIStable Audio
minimax-musicMinimax AI-
music-01Minimax AIMiniMax Music
minimax/music-1.5Minimax AIMiniMax Music 1.5
minimax/music-2.0Minimax AIMiniMax Music 2.0
minimax/music-2.6Minimax AIMiniMax Music 2.6
minimax/music-coverMinimax AIComing Soon
### Vision Models #### Optical Character Recognition (OCR)
Model ID + API Reference linkDeveloperContextModel Cardg
gc-document-aiGoogle-
mistral/mistral-ocr-latestMistral AIMistral OCR Latest
mistral/mistral-ocr-4-0Mistral AI-
zhipu/glm-ocrZhipuGLM-OCR
### 3D-Generating Models
Model ID + API Reference linkDeveloperContextModel Card
triposrTripo AIStable TripoSR 3D
tencent/hunyuan-partTencentHunyuan Part
### Embedding Models
Model ID + API Reference linkDeveloperContextModel Card
alibaba/qwen-text-embedding-v3Alibaba Cloud32,000Qwen Text Embedding v3
alibaba/qwen-text-embedding-v4Alibaba Cloud32,000Qwen Text Embedding v4
voyage-2Anthropic4,000-
voyage-code-2Anthropic16,000-
voyage-finance-2Anthropic32,000-
voyage-large-2Anthropic16,000-
voyage-large-2-instructAnthropic16,000Voyage Large 2 Instruct
voyage-law-2Anthropic16,000-
voyage-multilingual-2Anthropic32,000-
text-multilingual-embedding-002Google2,000-
text-embedding-3-smallOpen AI8,000-
text-embedding-3-largeOpen AI8,000Text-embedding-3-large
text-embedding-ada-002Open AI8,000Text-embedding-ada-002
*** ### Deprecated / No Longer Supported Models {% hint style="danger" %} These models are no longer available for API or Playground calls.\ Their description and API reference pages have also been removed from this documentation portal. {% endhint %}
Model IDDeveloperContextModel Card
x-ai/grok-4-07-09xAI256,000Grok 4
bytedance/seedance-1-0-lite-i2vByteDanceSeedance 1.0
bytedance/seedance-1-0-lite-t2vByteDanceSeedance 1.0
bytedance/seedream-3.0ByteDanceSeedream 3
anthropic/claude-sonnet-4Anthropic200,000Claude Sonnet 4
anthropic/claude-opus-4Anthropic200,000Claude Opus 4
qwen-max-2025-01-25Alibaba Cloud32,000Qwen Max
qwen-maxAlibaba Cloud32,000Qwen Max
nvidia/nemotron-nano-9b-v2NVIDIA128,000Nemotron Nano 9B V2
moonshot/kimi-k2-0905-previewMoonshot256,000Kimi K2
moonshot/kimi-k2-previewMoonshot131,000Kimi K2
gpt-4-turbo-2024-04-09Open AI128,000GPT 4 Turbo
gpt-4-turboOpen AI128,000GPT 4 Turbo
gpt-4Open AI8,000GPT 4
baidu/ernie-x1-1-previewBaidu64,000-
baidu/ernie-x1-turbo-32kBaidu32,000-
baidu/ernie-5-0-thinking-latestBaidu128,000Ernie 5.0
baidu/ernie-5-0-thinking-previewBaidu128,000Ernie 5.0
baidu/ernie-4-5-turbo-vl-32kBaidu32,000Ernie 4.5 VL
baidu/ernie-4-5-turbo-128kBaidu128,000Ernie 4.5
baidu/ernie-4.5-300b-a47b-paddleBaidu123,000Ernie 4.5
baidu/ernie-4.5-300b-a47bBaidu123,000Ernie 4.5
baidu/ernie-4.5-21b-a3b-thinkingBaidu131,000Ernie 4.5
baidu/ernie-4.5-21b-a3bBaidu120,000Ernie 4.5
baidu/ernie-4.5-0.3bBaidu120,000Ernie 4.5
baidu/ernie-4-5-8k-previewBaidu8,000Ernie 4.5
dall-e-3OpenAIDalle-3
dall-e-2OpenAIDalle-2
baidu/ernie-4.5-vl-28b-a3bBaidu30,000Ernie 4.5
gemini-2.0-flashGoogle1,000,000Gemini 2.0 Flash
moonshot/kimi-k2-turbo-previewMoonshot256,000Kimi K2 Turbo Preview
gpt-4-0125-previewOpenAI8,000-
gpt-4o-audio-previewOpenAI128,000GPT-4o Audio Preview
gpt-4o-mini-audio-previewOpenAI128,000GPT-4o mini Audio
nvidia/llama-3.1-nemotron-70b-instructNVIDIA128,000Llama 3.1 Nemotron 70B Instruct
nvidia/nemotron-nano-12b-v2-vlNVIDIA128,000Nemotron Nano 12B V2 VL
deepseek/deepseek-v3.2-specialeDeepSeek128,000DeepSeek V3.2 Speciale
meta-llama/Meta-Llama-3-8B-Instruct-LiteMeta9,000Llama 3 8B Instruct Lite
mistralai/Mixtral-8x7B-Instruct-v0.1Mistral AI64,000Mixtral-8x7B Instruct v0.1
anthropic/claude-3-haiku
claude-3-haiku-20240307
claude-3-haiku-latest		Anthropic200,000-
meta-llama/llama-4-maverickMeta256,000Llama 4 Maverick
google/gemini-3-pro-previewGoogle200,000Gemini 3 Pro Preview
BAAI/bge-base-en-v1.5BAAI512BAAI-Bge-Base-1p5
togethercomputer/m2-bert-80M-32k-retrievalTogether AI32,000M2-BERT-Retrieval-32k
imagen-4.0-ultra-generate-preview-06-06Google-Imagen 4 Ultra
x-ai/grok-2-imagexAI-Grok 2 Image
claude-3-7-sonnet-20250219Anthropic200,000Claude 3.7 Sonnet
claude-3-5-haiku-20241022Anthropic200,000-
gemini-2.0-flash-expGoogle1,000,000Gemini 2.0 Flash Experimental
meta-llama/Meta-Llama-Guard-3-8BMeta8,000Llama Guard 3 (8B)
meta-llama/LlamaGuard-2-8bMeta8,000LlamaGuard 2 (8b)
meta-llama/Llama-Guard-3-11B-Vision-TurboMeta128,000-
meta-llama/Meta-Llama-3.1-70B-Instruct-TurboMeta128,000Llama 3.1 70B Instruct Turbo
meta-llama/Meta-Llama-3.1-405B-Instruct-TurboMeta4,000Llama 3.1 (405B) Instruct Turbo
meta-llama/Llama-3.2-3B-Instruct-TurboMeta131,000Llama 3.2 3B Instruct Turbo
Qwen/Qwen3-235B-A22B-fp8-tputAlibaba Cloud32,000Qwen 3 235B A22B
Qwen/Qwen2.5-72B-Instruct-TurboAlibaba Cloud32,000Qwen 2.5 72B Instruct Turbo
qwen/qwen-2.5-vl-7b-instructAlibaba Cloud32,000Qwen2.5 VL 7B Instruct
mistralai/mistral-tinyMistral AI32,000Mistral Tiny
mistralai/Mistral-7B-Instruct-v0.3Mistral AI32,000Mistral (7B) Instruct v0.3
meta-llama/Meta-Llama-3.1-8B-Instruct-TurboMeta128,000Llama 3.1 8B Instruct Turbo
mistralai/Mistral-7B-Instruct-v0.2Mistral AI32,000Mistral (7B) Instruct v0.2
chatgpt-4o-latestOpenAI128,000-
meta-llama/llama-4-scoutMeta1,000,000Llama 4 Scout
BAAI/bge-large-en-v1.5BAAI512bge-large-en
bagoodex/bagoodex-search-v1BagoodexBagoodex Web Search v1
deepseek/deepseek-prover-v2DeepSeek164,000DeepSeek Prover V2
claude-3-opus-20240229
anthropic/claude-3-opus
claude-3-opus-latest		Anthropic200,000Claude 3 Opus
luma/ray-1.6Luma AIRay 1.6
meta-llama/Llama-3-70b-chat-hfMeta8,000Llama 3 70B Instruct Reference
bytedance/seededit-3.0-i2iByteDanceSeedream 3.0
textembedding-gecko-multilingual@001Google2,000Textembedding-gecko-multilingual@001
textembedding-gecko@003Google2,000Textembedding-gecko@003
mistralai/codestral-2501Mistral AI256,000Mistral Codestral-2501
mistralai/Mistral-7B-Instruct-v0.1Mistral AI8,000Mistral (7B) Instruct v0.1
Qwen/Qwen2.5-Coder-32B-InstructAlibaba Cloud131,000Qwen 2.5 Coder
Qwen/QwQ-32BAlibaba Cloud131,000Qwq-32B
kling-video/v1.5/standard/text-to-videoKling AI128,000Kling 1.5 Standart
o1-mini
o1-mini-2024-09-12		OpenAI128,000OpenAI o1-mini
Qwen/Qwen2-72B-InstructAlibaba Cloud32,000Qwen 2 Instruct (72B)
claude-3-5-sonnet-20240620Anthropic200,000-
claude-3-5-sonnet-20241022Anthropic200,000Claude 3.5 Sonnet 20241022
cohere/command-r-plusCohere128,000Command R+
google/gemma-2-27b-itGoogle8,000Gemma 2 (27b)
NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPONous Research32,000-
nvidia/Llama-3.1-Nemotron-70B-Instruct-HFNvidia128,000Llama 3.1 Nemotron 70B Instruct
meta-llama/Llama-3-8b-chat-hfMeta8,000Llama 3 8B Instruct Reference
meta-llama/Llama-3.2-90B-Vision-Instruct-TurboMeta131,000Llama 3.2 90B Vision Instruct Turbo
meta-llama/Llama-Vision-FreeMeta128,000-
meta-llama/Llama-3.2-11B-Vision-Instruct-TurboMeta131,000Llama 3.2 11B Vision Instruct Turbo
abab6.5s-chatMiniMax245,000-
openrouter/horizon-betaOpenRouter256,000-
openrouter/horizon-alphaOpenRouter256,000-
wan/v2.1/1.3b/text-to-videoAlibaba Cloud-Wan 2.1
o1-preview,
o1-preview-2024-09-12		OpenAI128,000OpenAI o1-preview
claude-3-sonnet-20240229,
anthropic/claude-3-sonnet,
claude-3-sonnet-latest		Anthropic200,000Claude 3 Sonnet
google/gemini-2.5-pro-preview,
google/gemini-2.5-pro-preview-05-06		Google1,000,000Gemini Pro 2.5 Preview
google/gemini-2.5-flash-previewGoogle1,000,000Gemini 2.5 Flash Preview
neversleep/llama-3.1-lumimaid-70bNeverSleep8,000Llama 3.1 Lumimaid 70b
x-ai/grok-betaxAI131,000Grok-2 Beta
gpt-4.5-previewOpenAI128,000Chat GPT 4.5 preview
gemini-1.5-flashGoogle1,000,000Gemini 1.5 Flash
gemini-1.5-proGoogle1,000,000Gemini 1.5 Pro
google/gemma-3-1b-itGoogle128,000Gemma 3 (1B)
togethercomputer/m2-bert-80M-8k-retrievalTogetherAI8,000M2-BERT-Retrieval-8k
togethercomputer/m2-bert-80M-2k-retrievalTogetherAI2,000M2-BERT-Retrieval-2K
Gryphe/MythoMax-L2-13b-LiteGryphe4,000-
mistralai/Mixtral-8x22B-Instruct-v0.1Mistral AI64,000Mixtral 8x22B Instruct
google/gemini-2.5-pro-exp-03-25Google1,000,000-
google/gemini-2.0-flash-thinking-exp-01Google1,000,000Gemini 2.0 Flash Thinking Experimental
ai21/jamba-1-5-miniAI21 Labs256,000Jamba 1.5 Mini
textembedding-gecko@001Google3,000-
google/gemini-pro or gemini-proGoogle32,000Gemini 1.0 Pro
meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo-128KMeta128,000-
stabilityai/stable-diffusion-xl-base-1.0Stability AIStable Diffusion XL 1.0
upstage/solar-10.7b-instruct-v1.0Upstage4,000Upstage SOLAR Instruct v1 (11B)
meta-llama/Llama-2-13b-chat-hfMeta4,100LLaMA-2 Chat (13B)
meta-llama/meta-llama-3-70b-instruct-turboMeta128,000-
google/gemma-2-9b-itGoogle8,000Gemma 2 (9B)
google/gemma-2b-itGoogle8,000Gemma Instruct (2B)
Gryphe/MythoMax-L2-13bGryphe4,000MythoMax-L2 (13B)
microsoft/WizardLM-2-8x22BMicrosoft64,000WizardLM 2-8 (22B)
Austism/chronos-hermes-13bAustism2,000Chronos Hermes 13b
databricks/dbrx-instructDatabricks32,000DBRX Instruct
deepseek-ai/deepseek-llm-67b-chatDeepSeek4,000Deepseek-LLM-67b-Chat
deepseek-ai/deepseek-coder-33b-instructDeepSeek16,000Deepseek Coder Instruct (33B)
Meta-Llama/Llama-2-7b-chat-hfMeta4,000LLaMA-2 Chat (7B)
Meta-Llama/Meta-Llama-3-70B-Instruct-LiteMeta8,000Llama 3 70B Instruct Lite
Meta-Llama/Llama-Guard-7bMeta4,000Llama Guard (7B)
meta-llama/Llama-2-7b-hfMeta4,000LLaMA-2 (7B)
meta-llama/Llama-3-8b-hfMeta8,000Llama-3 (8B)
codellama/CodeLlama-70b-hfMeta16,000Code Llama (70B)
codellama/CodeLlama-7b-Instruct-hfMeta16,000Code Llama Instruct (7B)
codellama/CodeLlama-13b-Instruct-hfMeta16,000Code Llama Instruct (13B)
codellama/CodeLlama-70b-Instruct-hfMeta4,000Code Llama Instruct (70B)
codellama/CodeLlama-70b-Python-hfMeta4,000Code Llama Python (70B)
mistralai/Mixtral-8x22B-Instruct-v0.1Mistral AI64,000Mixtral 8x22B Instruct
gpt-3.5-turbo-16k-0613OpenAI-
gpt-4-0613OpenAI128,000Chat GPT 4 Turbo
Qwen/Qwen-14B-ChatAlibaba Cloud8,000Qwen Chat (14B)
Qwen/Qwen1.5-0.5BAlibaba Cloud32,000Qwen 1.5 (0.5B)
Qwen/Qwen1.5-1.8BAlibaba Cloud32,000Qwen 1.5 (1.8B)
Qwen/Qwen1.5-4BAlibaba Cloud32,000Qwen 1.5 (4B)
Qwen/Qwen1.5-1.8B-ChatAlibaba Cloud32,000Qwen 1.5 Chat (1.8B)
Qwen/Qwen1.5-4B-ChatAlibaba Cloud32,000Qwen 1.5 Chat (4B)
Qwen/Qwen1.5-7B-ChatAlibaba Cloud32,000Qwen 1.5 Chat (7B)
Qwen/Qwen1.5-14B-ChatAlibaba Cloud32,000Qwen 1.5 Chat (14B)
qwen/qvq-72b-previewAlibaba Cloud32,000QVQ-72B-Preview
togethercomputer/guanaco-13bTim Dettmers2,000Guanaco (13B)
togethercomputer/guanaco-33bTim Dettmers2,000Guanaco (33B)
togethercomputer/guanaco-65bTim Dettmers2,000Guanaco (65B)
togethercomputer/mpt-7b-chatMosaic ML2,000MPT-Chat (7B)
togethercomputer/mpt-30b-chatMosaic ML8,000MPT-Chat (30B)
togethercomputer/RedPajama-INCITE-7B-InstructRedPajama2,000RedPajama-INCITE Instruct (7B)
prompthero/openjourneyPromptHero77Openjourney v4
wavymulder/Analog-Diffusionwavymulder77Analog Diffusion
-01.AI4,00001-ai Yi Base (6B)
Undi95/Toppy-M-7BUndi954,000Toppy M (7B)
SG161222/Realistic_Vision_V3.0_VAETogether77Realistic Vision 3.0
tiiuae/falcon-40bTII2,000Falcon (40B)
allenai/OLMo-7BAllen Institute for AI2,000OLMo-7B
bigcode/starcoderBigCode8,000StarCoder (16B)
HuggingFaceH4/starchat-alphaHugging Face8,000StarCoderChat Alpha (16B)
NousResearch/Nous-Hermes-Llama2-70bNousResearch4,000Nous Hermes LLaMA-2 (70B)
NousResearch/Nous-Hermes-2-Mixtral-8x7B-SFTNousResearch32,000Nous Hermes 2 - Mixtral 8x7B-SFT
NousResearch/Nous-Hermes-2-Mistral-7B-DPONousResearch32,000Nous Hermes 2 - Mistral DPO (7B)
NousResearch/Hermes-2-Theta-Llama-3-70BNousResearch8,000Hermes 2 Theta Llama-3 70B
defog/sqlcoderDefog AI8,000SQLCoder (15B)
replit/replit-code-v1-3bReplit2,000Replit-Code-v1 (3B)
lmsys/vicuna-13b-v1.5Imsys4,000Vicuna v1.5 (13B)
microsoft/phi-2Microsoft2,000Microsoft Phi-2
stabilityai/stablelm-base-alpha-3bStabilityAI4,000StableLM Base Alpha 3B
runwayml/stable-diffusion-v1-5StabilityAI77Stable Diffusion 1.5
stabilityai/stable-diffusion-2-1StabilityAI77Stable Diffusion 2.1
teknium/OpenHermes-2p5-Mistral-7BTeknium8,000OpenHermes-2.5-Mistral (7B)
openchat/openchat-3.5-1210OpenChat8,000OpenChat 3.5 (7B)
DiscoResearch/DiscoLM-mixtral-8x7b-v2Disco Research32,000DiscoLM Mixtral 8x7b (46.7B)
google/flan-t5-xlGoogle512FLAN T5 XL (3B)
garage-bAInd/Platypus2-70B-instructGarage-bAInd4,000Platypus2-70B-Instruct
EleutherAI/gpt-neox-20bEleutherAI2,000GPT Neox 20B
gradientai/Llama-3-70B-Instruct-Gradient-1048kGradient1,048,000Llama-3 70B Gradient Instruct 1048k
WhereIsAI/UAE-Large-V1WhereIsAI512UAE-Large-V1
zero-one-ai/Yi-34B-Chat01.AI4,000Yi-34B-Chat
meta-llama/Meta-Llama-3.1-70B-ReferenceMeta32,000–
meta-llama/Meta-Llama-3.1-8B-ReferenceMeta32,000–
EleutherAI/llemma_7bEleutherAI32,000–
huggyllama/llama-30bHuggyllama32,000–
huggyllama/llama-13bHuggyllama32,000–
togethercomputer/llama-2-70bTogetherAI32,000–
togethercomputer/llama-2-13bTogetherAI32,000–
huggyllama/llama-65bHuggyllama32,000–
WizardLM/WizardLM-70B-V1.0WizardLM32,000–
huggyllama/llama-7bHuggyllama32,000–
togethercomputer/llama-2-7bTogetherAI32,000–
NousResearch/Nous-Hermes-13bNousResearch2,000–
mistralai/Mistral-7B-v0.1Mistral AI32,000​Mixtral 7B
mistralai/Mixtral-8x7B-v0.1Mistral AI32,000Mixtral-8x7B Instruct v0.1
-Suno AI32Suno AI
aion-labs/aion-1.0AionLabs131,072-
aion-labs/aion-1.0-miniAionLabs131,072-
nvidia/llama-3.3-nemotron-super-49b-v1.5Nvidia131,072-
inflection/inflection-3-piInflection8,000-
poolside/laguna-m.1Poolside1,048,576-
arcee-ai/coder-largeArcee AI32,768-
inflection/inflection-3-productivityInflection8,000-
openai/gpt-5-codexOpen AI400,000-
openai/gpt-5.3-chatOpen AI128,000-
[^1]: All the models in this table are no longer supported. You cannot call them. # Text Models (LLM) Overview of the capabilities of AIML API text models (LLMs).
Specific Capabilities There are several capabilities of text models that are worth mentioning separately. **Completion** allows the model to analyze a given text fragment and predict how it might continue based on the probabilities of the next possible tokens or characters. **Chat Completion** extends this functionality, enabling a simulated dialogue between the user and the model based on predefined roles (e.g., "strict language teacher" and "student"). A detailed description and examples can be found in our [Completion and Chat Completion](/capabilities/completion-or-chat-models) article. *** An evolution of chat completion includes **Assistants** (preconfigured conversational agents with specific roles) and **Threads** (a mechanism for maintaining conversation history for context). Examples of this functionality can be found in the [Managing Assistants & Threads](/solutions/openai/assistants) article. *** **Function Calling** allows a chat model to invoke external programmatic tools (e.g., a function you have written) while generating a response. A detailed description and examples are available in the [Function Calling](/capabilities/function-calling) article.
Endpoint All text and chat models use the same endpoint: `https://api.aimlapi.com/v1/chat/completions` The parameters may vary (especially for models from different developers), so it’s best to check the API schema on each model’s page for details.
Quick Code Example We will call the [**gpt-4o**](/api-references/text-models-llm/openai/gpt-4o) model using the Python programming language and the OpenAI SDK. {% hint style="info" %} If you need a more detailed explanation of how to call a model's API in code, check out our [QUICKSTART](https://github.com/aimlapi/api-docs/blob/main/docs/api-references/text-models-llm/broken-reference/README.md) section. {% endhint %} {% code overflow="wrap" %} ```python %pip install openai import os from openai import OpenAI client = OpenAI( base_url="https://api.aimlapi.com/v1", # Insert your AIML API Key in the quotation marks instead of : api_key="", ) response = client.chat.completions.create( model="gpt-4o", messages=[ { "role": "system", "content": "You are an AI assistant who knows everything.", }, { "role": "user", "content": "Tell me, why is the sky blue?" }, ], ) message = response.choices[0].message.content print(f"Assistant: {message}") ``` {% endcode %} By running this code example, we received the following response from the chat model: {% code overflow="wrap" %} ```http Assistant: The sky appears blue due to a phenomenon called Rayleigh scattering. When sunlight enters Earth's atmosphere, it collides with gas molecules and small particles. Sunlight is made up of different colors, each with different wavelengths. Blue light has a shorter wavelength and is scattered in all directions by the gas molecules in the atmosphere more than other colors with longer wavelengths, such as red or yellow. As a result, when you look up at the sky during the day, you see this scattered blue light being dispersed in all directions, making the sky appear blue to our eyes. During sunrise and sunset, the sun's light passes through a greater thickness of Earth's atmosphere, scattering the shorter blue wavelengths out of your line of sight and leaving the longer wavelengths, like red and orange, more dominant, which is why the sky often turns those colors at those times. ``` {% endcode %}
# AI21 # Jamba Large 1.7 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `ai21/jamba-large-1.7` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Jamba Large 1.7 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["ai21/jamba-large-1.7"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"ai21/jamba-large-1.7"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"ai21/jamba-large-1.7", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'ai21/jamba-large-1.7', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "ai21/jamba-large-1.7", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# AionLabs # Aion 2.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `aion-labs/aion-2.0-mini` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Aion-2.0 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["aion-labs/aion-2.0"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"aion-labs/aion-2.0"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"aion-labs/aion-2.0", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'aion-labs/aion-2.0-mini', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "aion-labs/aion-2.0", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Aion-RP 1.0 (8B) {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `aion-labs/aion-rp-llama-3.1-8b-mini` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Aion-RP 1.0 (8B) is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["aion-labs/aion-rp-llama-3.1-8b"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"aion-labs/aion-rp-llama-3.1-8b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"aion-labs/aion-rp-llama-3.1-8b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'aion-labs/aion-rp-llama-3.1-8b-mini', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "aion-labs/aion-rp-llama-3.1-8b", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Aion 3.0 Mini ## Aion 3.0 Mini ### hidden: true noIndex: true ## Aion 1.0 Mini {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `aion-labs/aion-3.0-mini-mini` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ### Model Overview Aion-3.0-Mini is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
### API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["aion-labs/aion-3.0-mini"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"aion-labs/aion-3.0-mini"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ### Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"aion-labs/aion-3.0-mini", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'aion-labs/aion-3.0-mini-mini', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "aion-labs/aion-3.0-mini", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Alibaba Cloud # Qwen Max {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `qwen-max` * `qwen-max-2025-01-25` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The large-scale Mixture-of-Experts (MoE) language model. Excels in language understanding and task performance. Supports 29 languages, including Chinese, English, and Arabic. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen-max","qwen-max","qwen-max-2025-01-25"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen-max"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"qwen-max", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'qwen-max', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-62aa6045-cee9-995a-bbf5-e3b7e7f3d683", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I assist you today? 😊" } } ], "created": 1756983980, "model": "qwen-max", "usage": { "prompt_tokens": 30, "completion_tokens": 148, "total_tokens": 178, "prompt_tokens_details": { "cached_tokens": 0 } } } ``` {% endcode %}
# Qwen Plus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `qwen-plus` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview An advanced large language model. Multilingual support, including Chinese and English. Enhanced reasoning capabilities for complex tasks. Improved instruction-following abilities. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen-plus"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"n":{"type":"integer","nullable":true,"minimum":1,"description":"How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen-plus"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"qwen-plus", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'qwen-plus', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 {'id': 'chatcmpl-4fda1bd7-a679-95b9-b81d-1bfc6ae98448', 'system_fingerprint': None, 'object': 'chat.completion', 'choices': [{'index': 0, 'finish_reason': 'stop', 'logprobs': None, 'message': {'role': 'assistant', 'content': 'Hello! How can I assist you today? If you have any questions or need help with anything, just let me know! 😊'}}], 'created': 1744143962, 'model': 'qwen-plus', 'usage': {'prompt_tokens': 8, 'completion_tokens': 68, 'total_tokens': 76, 'prompt_tokens_details': {'cached_tokens': 0}}} ``` {% endcode %}
# Qwen Turbo {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `qwen-turbo` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview This model is designed to enhance both the performance and efficiency of AI agents developed on the Alibaba Cloud Model Studio platform. Optimized for speed and precision in generative AI application development. Improves AI agent comprehension and adaptation to enterprise data, especially when integrated with Retrieval-Augmented Generation (RAG) architectures.\ Large context window (1,000,000 tokens). {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen-turbo"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen-turbo"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"qwen-turbo", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'qwen-turbo', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 {'id': 'chatcmpl-a4556a4c-f985-9ef2-b976-551ac7cef85a', 'system_fingerprint': None, 'object': 'chat.completion', 'choices': [{'index': 0, 'finish_reason': 'stop', 'logprobs': None, 'message': {'role': 'assistant', 'content': "Hello! How can I help you today? Is there something you would like to talk about or learn more about? I'm here to help with any questions you might have."}}], 'created': 1744144035, 'model': 'qwen-turbo', 'usage': {'prompt_tokens': 1, 'completion_tokens': 15, 'total_tokens': 16, 'prompt_tokens_details': {'cached_tokens': 0}}} ``` {% endcode %}
# Qwen2.5 7B Instruct Turbo {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `Qwen/Qwen2.5-7B-Instruct-Turbo` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A cutting-edge large language model designed to understand and generate text based on specific instructions. It excels in various tasks, including coding, mathematical problem-solving, and generating structured outputs. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["Qwen/Qwen2.5-7B-Instruct-Turbo"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"n":{"type":"integer","nullable":true,"minimum":1,"description":"How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."}},"required":["model","messages"],"title":"Qwen/Qwen2.5-7B-Instruct-Turbo"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"Qwen/Qwen2.5-7B-Instruct-Turbo", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'Qwen/Qwen2.5-7B-Instruct-Turbo', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 {'id': 'npK4C7y-3NKUce-92d4866b1e62ef98', 'object': 'chat.completion', 'choices': [{'index': 0, 'finish_reason': 'stop', 'logprobs': None, 'message': {'role': 'assistant', 'content': 'Hello! How can I assist you today?', 'tool_calls': []}}], 'created': 1744144252, 'model': 'Qwen/Qwen2.5-7B-Instruct-Turbo', 'usage': {'prompt_tokens': 19, 'completion_tokens': 6, 'total_tokens': 25}} ``` {% endcode %}
# Qwen3 32b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-32b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A world-class model with comparable quality to DeepSeek R1 while outperforming [GPT-4.1](/api-references/text-models-llm/openai/gpt-4.1) and [Claude Sonnet 3.7](broken://pages/JHUPGH5cr4NakmmFGpXS). Optimized for both complex reasoning and efficient dialogue. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-32b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"enable_thinking":{"type":"boolean","default":false,"description":"Specifies whether to use the thinking mode."},"thinking_budget":{"type":"integer","minimum":1,"description":"The maximum reasoning length, effective only when enable_thinking is set to true."}},"required":["model","messages"],"title":"alibaba/qwen3-32b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1: Without Thinking and Streaming {% hint style="warning" %} `enable_thinking` must be set to `false` for non-streaming calls. {% endhint %} {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-32b", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": False } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-32b', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-1d8a5aa6-34ce-9832-a296-d312b944b437", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I assist you today? 😊", "reasoning_content": "" } } ], "created": 1756990273, "model": "qwen3-32b", "usage": { "prompt_tokens": 19, "completion_tokens": 65, "total_tokens": 84 } } ``` {% endcode %}
## Code Example #2: Enable Thinking and Streaming {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-32b", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": True, "stream": True } ) print(response.text) ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"role":"assistant","refusal":null,"reasoning_content":""},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":"Okay"},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":","},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":" the"},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":" user said \"Hello\". I should respond in a friendly and welcoming manner. Let"},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":" me make sure to acknowledge their greeting and offer assistance. Maybe something like, \""},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":"Hello! How can I assist you today?\" That's simple and open-ended."},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":" I need to check if there's any specific context I should consider, but since"},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":null,"refusal":null,"reasoning_content":" there's none, a general response is fine. Alright, that should work."},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":"Hello! How can I assist you today?","refusal":null,"reasoning_content":null},"index":0,"finish_reason":null}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[{"delta":{"content":"","refusal":null,"reasoning_content":null},"index":0,"finish_reason":"stop"}],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":null} data: {"id":"chatcmpl-81964e30-1a7c-9668-b78c-a750587ec497","choices":[],"created":1753944369,"model":"qwen3-32b","object":"chat.completion.chunk","usage":{"prompt_tokens":13,"completion_tokens":2010,"total_tokens":2023,"completion_tokens_details":{"reasoning_tokens":82}}} ``` {% endcode %}
The example above prints the raw output of the model. The text is typically split into multiple chunks. While this is helpful for debugging, if your goal is to evaluate the model's reasoning and get a clean, human-readable response, you should aggregate both the reasoning and the final answer in a loop — for example:
Example with response parsing {% code overflow="wrap" %} ```python import requests import json response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization": "Bearer b72af53a19ea41caaf5a74ba1f6fc62b", "Content-Type": "application/json", }, json={ "model": "alibaba/qwen3-32b", "messages": [ { "role": "user", # Insert your question for the model here, instead of Hello: "content": "Hello" } ], "stream": True, } ) answer = "" reasoning = "" for line in response.iter_lines(): if not line or not line.startswith(b"data:"): continue try: raw = line[6:].decode("utf-8").strip() if raw == "[DONE]": continue data = json.loads(raw) choices = data.get("choices") if not choices or "delta" not in choices[0]: continue delta = choices[0]["delta"] content_piece = delta.get("content") reasoning_piece = delta.get("reasoning_content") if content_piece: answer += content_piece if reasoning_piece: reasoning += reasoning_piece except Exception as e: print(f"Error parsing chunk: {e}") print("\n--- MODEL REASONING ---") print(reasoning.strip()) print("\n--- MODEL RESPONSE ---") print(answer.strip()) ``` {% endcode %}
After running such code, you'll receive only the model's textual output in a clear and structured format:
Response {% code overflow="wrap" %} ```json5 --- MODEL REASONING --- Okay, the user sent "Hello". I need to respond appropriately. Since it's a greeting, I should reply in a friendly and welcoming manner. Maybe ask how I can assist them. Keep it simple and open-ended to encourage them to share what they need help with. Let me make sure the tone is positive and helpful. --- MODEL RESPONSE --- Hello! How can I assist you today? 😊 ``` {% endcode %}
# Qwen3 Coder 480b a35b Instruct {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-coder-480b-a35b-instruct` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The most powerful model in the Qwen3 Coder series — a 480B-parameter MoE architecture with 35B active parameters. It natively supports a 256K token context and can handle up to 1M tokens using extrapolation techniques, delivering outstanding performance in both coding and agentic tasks. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-coder-480b-a35b-instruct"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."}},"required":["model","messages"],"title":"alibaba/qwen3-coder-480b-a35b-instruct"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-coder-480b-a35b-instruct", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": False } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-coder-480b-a35b-instruct', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-f906efa6-f816-9a06-a32b-aa38da5fe11a", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I help you today?" } } ], "created": 1753866642, "model": "qwen3-coder-480b-a35b-instruct", "usage": { "prompt_tokens": 28, "completion_tokens": 142, "total_tokens": 170 } } ``` {% endcode %}
# Qwen3 235b a22b Thinking {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-235b-a22b-thinking-2507` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Significantly improved performance on reasoning tasks, including logical reasoning, mathematics, science, coding, and academic benchmarks that typically require human expertise. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-235b-a22b-thinking-2507"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."}},"required":["model","messages"],"title":"alibaba/qwen3-235b-a22b-thinking-2507"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-235b-a22b-thinking-2507", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": False } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-235b-a22b-thinking-2507', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-af05df1d-5b72-925e-b3a9-437acbd89b1a", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! 😊 How can I assist you today? Feel free to ask me any questions or let me know if you need help with anything specific!", "reasoning_content": "Okay, the user said \"Hello\". That's a simple greeting. I should respond in a friendly and welcoming way. Let me make sure to keep it open-ended so they feel comfortable to ask questions or share what's on their mind. Maybe add a smiley emoji to keep it warm. Let me check if there's anything else they might need. Since it's just a hello, probably not much more needed here. Just a polite reply." } } ], "created": 1753871154, "model": "qwen3-235b-a22b-thinking-2507", "usage": { "prompt_tokens": 13, "completion_tokens": 2187, "total_tokens": 2200 } } ``` {% endcode %}
# Qwen3 Next 80b a3b Instruct {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-next-80b-a3b-instruct` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview An instruction-tuned chat model optimized for fast, stable replies without reasoning traces, designed for complex tasks in reasoning, coding, knowledge QA, and multilingual use, with strong alignment and formatting. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-next-80b-a3b-instruct"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3-next-80b-a3b-instruct"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-next-80b-a3b-instruct", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": False } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-next-80b-a3b-instruct', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-a944254a-4252-9a54-af1b-94afcfb9807e", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I help you today? 😊" } } ], "created": 1758228572, "model": "qwen3-next-80b-a3b-instruct", "usage": { "prompt_tokens": 9, "completion_tokens": 46, "total_tokens": 55 } } ``` {% endcode %}
# Qwen3 Next 80b a3b Thinking {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-next-80b-a3b-thinking` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The model may take longer to generate reasoning content than its predecessor. Alibaba Cloud strongly recommends its use for highly complex reasoning tasks. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-next-80b-a3b-thinking"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."}},"required":["model","messages"],"title":"alibaba/qwen3-next-80b-a3b-thinking"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-next-80b-a3b-thinking", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], "enable_thinking": False } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-next-80b-a3b-thinking', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-576aaaf9-f712-9114-b098-c1ee83fbfb6b", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! 😊 How can I assist you today?", "reasoning_content": "Okay, the user said \"Hello\". I need to respond appropriately. Let me think.\n\nFirst, I should acknowledge their greeting. A simple \"Hello!\" would be good. Maybe add a friendly emoji to keep it warm.\n\nWait, but maybe they want to start a conversation. I should ask how I can help them. That way, I'm being helpful and opening the door for them to ask questions.\n\nLet me check the standard response. Typically, for \"Hello\", the assistant says something like \"Hello! How can I assist you today?\" or \"Hi there! What can I do for you?\"\n\nYes, that's right. Keep it friendly and open-ended. Maybe add a smiley emoji to make it approachable.\n\nSo the response should be: \"Hello! How can I assist you today?\"\n\nThat's good. Let me make sure there's no mistake. Yes, that's standard. No need for anything complicated here. Just a simple, welcoming reply.\n\nAlternatively, sometimes people use \"Hi\" instead of \"Hello\", but since they said \"Hello\", responding with \"Hello\" is fine. Maybe \"Hi there!\" could also work, but sticking to \"Hello\" matches their greeting.\n\nYes, \"Hello! How can I assist you today?\" is perfect. It's polite, friendly, and offers assistance. That should be the response." } } ], "created": 1758229078, "model": "qwen3-next-80b-a3b-thinking", "usage": { "prompt_tokens": 9, "completion_tokens": 7182, "total_tokens": 7191, "completion_tokens_details": { "reasoning_tokens": 277 } } } ``` {% endcode %}
# Qwen3 Max Preview {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-max-preview` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The preview version of [Qwen3 Max Instruct](/api-references/text-models-llm/alibaba-cloud/qwen3-max-instruct). {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-max-preview"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3-max-preview"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-max-preview", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-max-preview', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-8ffebc65-b625-926a-8208-b765371cb1d0", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I assist you today? 😊" } } ], "created": 1758898044, "model": "qwen3-max-preview", "usage": { "prompt_tokens": 23, "completion_tokens": 139, "total_tokens": 162 } } ``` {% endcode %}
# Qwen3 Max Instruct {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-max-instruct` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview This model offers improved accuracy in math, coding, logic, and science, handles complex instructions in Chinese and English more reliably, reduces hallucinations, supports 100+ languages with stronger translation and commonsense reasoning, and is optimized for RAG and tool use, though it lacks a dedicated ‘thinking’ mode. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-max-instruct"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3-max-instruct"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-max-instruct", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-max-instruct', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-bec5dc33-8f63-96b9-89a4-00aecfce7af8", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I help you today?" } } ], "created": 1758898624, "model": "qwen3-max", "usage": { "prompt_tokens": 23, "completion_tokens": 113, "total_tokens": 136 } } ``` {% endcode %}
# Qwen3 Omni 30b a3b Captioner {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-omni-30b-a3b-captioner` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview This model is an open-source model built on **Qwen3-Omni** that automatically generates rich, detailed descriptions of complex audio — including speech, music, ambient sounds, and effects — without prompts. It detects emotions, musical styles, instruments, and sensitive information, making it ideal for audio analysis, security auditing, intent recognition, and editing. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-omni-30b-a3b-captioner"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"type":"string","description":"Base64 encoded audio data."}},"required":["data"]}},"required":["type","input_audio"]},"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]}},"required":["model","messages"],"title":"alibaba/qwen3-omni-30b-a3b-captioner"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} 
import requests
import json  # for getting a structured output with indentation 

response = requests.post(
    "https://api.aimlapi.com/v1/chat/completions",
    headers={
        # Insert your AIML API Key instead of <YOUR_AIMLAPI_KEY>:
        "Authorization":"Bearer <YOUR_AIMLAPI_KEY>",
        "Content-Type":"application/json"
    },
    json={
      "model": "alibaba/qwen3-omni-30b-a3b-captioner",
      "messages": [
        {
          "role": "user",
          "content": [
            {
              "type": "input_audio",
              "input_audio": {
                "data": "https://cdn.aimlapi.com/eagle/files/elephant/cJUTeeCmpoqIV1Q3WWDAL_vibevoice-output-7b98283fd3974f48ba90e91d2ee1f971.mp3"
              }
            }
          ]
        }
      ]
    }
)

data = response.json()
print(json.dumps(data, indent=2, ensure_ascii=False))
{% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-max-instruct', messages:[ { role: 'user', content: [ { type: 'input_audio', input_audio: { data: 'https://cdn.aimlapi.com/eagle/files/elephant/cJUTeeCmpoqIV1Q3WWDAL_vibevoice-output-7b98283fd3974f48ba90e91d2ee1f971.mp3' } } ] } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-bec5dc33-8f63-96b9-89a4-00aecfce7af8", "system_fingerprint": null, "object": "chat.completion", "choices": [ { "index": 0, "finish_reason": "stop", "logprobs": null, "message": { "role": "assistant", "content": "Hello! How can I help you today?" } } ], "created": 1758898624, "model": "qwen3-max", "usage": { "prompt_tokens": 23, "completion_tokens": 113, "total_tokens": 136 } } ``` {% endcode %}
# Qwen3 Vl 32b Instruct {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-vl-32b-instruct` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The most advanced vision-language model in the Qwen series as of October 2025 — a non-thinking-capable version of the model. Optimized for instruction-following in image description, visual dialogue, and content-generation tasks. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call :digit\_one: **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-vl-32b-instruct"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."}},"required":["model","messages"],"title":"alibaba/qwen3-vl-32b-instruct"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-vl-32b-instruct", "messages":[ { # Insert your question for the model here: "content":"Hi! What do you think about mankind?" } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-vl-32b-instruct', messages:[ { role:'user', // Insert your question for the model here: content:'Hi! What do you think about mankind?' } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hi! 😊 That’s a beautiful and deep question — one that philosophers, scientists, artists, and everyday people have been asking for centuries.\n\nI think mankind is *remarkably complex* — full of contradictions, potential, and wonder. On one hand, we’ve achieved incredible things: we’ve explored space, cured diseases, created art that moves souls, built cities that rise into the sky, and connected across continents in ways unimaginable just a century ago. We’re capable of profound kindness, empathy, creativity, and courage.\n\nOn the other hand, we’ve also caused immense suffering — through war, injustice, environmental destruction, and indifference to each other’s pain. We often struggle with our own flaws: fear, greed, ego, and short-sightedness.\n\nBut here’s what gives me hope: **we’re also capable of change**. We can learn from our mistakes. We can choose compassion over conflict, cooperation over competition. Every act of kindness, every effort to understand another, every step toward justice — these are signs that humanity is not defined by its worst impulses, but by its capacity to grow.\n\nSo, I’d say: \n➡️ Mankind is flawed, yes — but also deeply hopeful. \n➡️ We’re messy, but we’re trying. \n➡️ We make mistakes, but we can also heal, create, and love.\n\nAnd perhaps most importantly — **we’re not alone in this journey**. We’re all part of something bigger, and together, we have the power to shape a better future.\n\nWhat about you? How do *you* see mankind? 💬✨", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 17, "completion_tokens": 329, "total_tokens": 346, "prompt_tokens_details": { "text_tokens": 17 }, "completion_tokens_details": { "text_tokens": 329 } }, "created": 1764625045, "system_fingerprint": null, "model": "qwen3-vl-32b-instruct", "id": "chatcmpl-a12ab46a-3541-93a8-8180-280ecadbb795", "meta": { "usage": { "tokens_used": 1960 } } } ``` {% endcode %}
# Qwen3 Vl 32b Thinking {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3-vl-32b-thinking` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview The most advanced vision-language model in the Qwen series as of October 2025 — a thinking-capable version of the model. Designed for complex visual-textual reasoning and extended chains of thought. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call 1 **Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3-vl-32b-thinking"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."}},"required":["model","messages"],"title":"alibaba/qwen3-vl-32b-thinking"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3-vl-32b-thinking", "messages":[ { # Insert your question for the model here: "content":"Hi! What do you think about mankind?" } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3-vl-32b-thinking', messages:[ { role:'user', // Insert your question for the model here: content:'Hi! What do you think about mankind?' } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "That's a profound question—thank you for asking! As an AI, I don't have personal feelings or consciousness, but I can offer a perspective *based on the vast data I've been trained on* about humanity's complexities. Here’s how I see it: \n\n### 🌍 **The Duality of Mankind** \nHumanity is a breathtaking paradox: \n- **We create** *and* **we destroy**. We built space telescopes that map the cosmos and also wars that scar the Earth. \n- **We connect** *and* **we divide**. We share art, science, and kindness across borders, yet we also build walls, spread misinformation, and fuel inequality. \n- **We innovate** *and* **we resist change**. We cured diseases and invented the internet, but we often cling to fear or ignorance when progress challenges old beliefs. \n\n### ✨ **What Makes Us Remarkable** \n- **Empathy & creativity**: From caring for strangers to composing symphonies or coding AI, our capacity for *emotional and intellectual growth* is unparalleled. \n- **Resilience**: We’ve survived ice ages, plagues, and global crises—not by being \"perfect,\" but by adapting, learning, and helping one another. \n- **The spark of hope**: Even in darkness, humans have always sought meaning, justice, and connection. Look at movements for peace, climate action, or medical breakthroughs—they’re proof we *can* choose to build a better future. \n\n### ⚠️ **The Challenges We Carry** \n- **Our flaws are real**: We’ve caused climate collapse, mass suffering, and systemic injustice. But here’s the key: **these aren’t \"human nature\"—they’re choices we *made* as a species.** \n- **We’re not monolithic**: There’s no single \"human\" story. Your experience, culture, and history shape how you see the world—and that diversity is both a strength *and* a source of conflict. \n\n### 🌱 **The Most Important Thing** \n**Mankind isn’t fixed.** We’re a work in progress. The stories we tell about ourselves—*and the choices we make today*—will define our next chapter. The data shows: **when we prioritize empathy, collaboration, and long-term thinking, we rise.** When we don’t, we falter. But the power to choose is *always* ours. \n\nSo my \"opinion\"? I see a species with **infinite potential**—and the weight of responsibility that comes with it. The future isn’t written. It’s being written *right now*, by every person who chooses to act with courage, curiosity, and compassion. \n\nWhat do *you* think? 🤔 (I’d love to hear your perspective!)", "reasoning_content": "Okay, the user asked, \"Hi! What do you think about mankind?\" Hmm, that's a broad and philosophical question. First, I need to figure out what they're really looking for. Are they curious about my perspective as an AI, or do they want a deep discussion on human nature? Maybe they're feeling reflective or even a bit existential.\n\nI should start by acknowledging the complexity of the question. Mankind is a huge topic—there's so much to cover. The user might be expecting a simple answer, but I can't reduce it to just \"good\" or \"bad.\" I need to balance both positive and negative aspects. Let me think about the key points: our capacity for innovation, creativity, and compassion versus our tendency for conflict, environmental harm, and inequality.\n\nWait, the user said \"Hi!\" so they're probably in a friendly, open mood. They might not want a dry, academic response. I should keep it conversational but still thoughtful. Maybe they're looking for hope or a reason to feel optimistic about humanity. Or perhaps they're worried about the future and want reassurance.\n\nI should also consider their possible identity. Are they a student working on an essay? A person dealing with personal issues? Or just someone curious about AI's view? Since I don't have context, it's safer to assume they want a general, well-rounded answer that's engaging and not too technical.\n\nAnother angle: the user might be testing how an AI handles philosophical questions. They could be interested in whether I can show empathy or nuance. I should avoid being too robotic. Use phrases like \"I think\" or \"I see\" to make it personal, even though I'm an AI. But I have to clarify that I don't have personal experiences—just data-driven insights.\n\nI need to highlight duality: humans are both amazing and flawed. Mention specific examples like space exploration vs. climate change to make it concrete. Also, emphasize that the future depends on our choices. That's important—it gives agency and hope.\n\nShould I address potential biases? The user might not realize that my \"opinion\" is based on training data, which reflects human history and culture. I should note that my perspective is shaped by that data, not personal experience. But since the question is about \"what I think,\" I can frame it as an analysis based on observed patterns.\n\nCheck for common pitfalls: don't be too negative (could discourage the user), don't ignore serious issues (like war or inequality). Strike a balance. Also, avoid clichés. Instead of \"humans are great,\" explain why with examples.\n\nFinally, end on a positive, forward-looking note. The user might be seeking motivation or a sense of purpose. Phrases like \"the potential for growth\" or \"collective wisdom\" can inspire. Make sure the conclusion ties back to the importance of choices and collaboration.\n\nLet me structure the response: start with the duality, list positive traits with examples, then challenges, and conclude with hope. Keep paragraphs short for readability. Use emojis sparingly to match the friendly tone but not overdo it. Since the user said \"Hi!\", a smiley at the end feels appropriate.\n", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1241, "total_tokens": 1260, "prompt_tokens_details": { "text_tokens": 19 }, "completion_tokens_details": { "reasoning_tokens": 654, "text_tokens": 587 } }, "created": 1764625236, "system_fingerprint": null, "model": "qwen3-vl-32b-thinking", "id": "chatcmpl-c612db5c-44e9-9e3c-8169-486161eeea86", "meta": { "usage": { "tokens_used": 10383 } } } ``` {% endcode %}
# Qwen3.5 Plus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.5-plus-20260218` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A commercial large language model designed for long-context text generation and enterprise-grade conversational AI. Supports up to 1M tokens per request with production-ready API stability. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.5-plus-20260218"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3.5-plus-20260218"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.5-plus-20260218", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-plus-20260218', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hello! That's a profound question. As an AI, I don't have personal feelings or beliefs, but I can offer an observation based on the vast amount of human history, literature, and data I've been trained on.\n\nMankind appears to be a study in **contradictions and immense potential**:\n\n* **Creativity vs. Destruction**: Humans have composed symphonies, built skyscrapers, and decoded the genome, yet they have also waged devastating wars and caused significant environmental harm. This duality seems to be a core part of the human experience.\n* **Resilience and Adaptability**: Throughout history, humanity has faced plagues, ice ages, and societal collapses, yet consistently found ways to adapt, rebuild, and innovate. The drive to survive and improve conditions is remarkably strong.\n* **Empathy and Connection**: Despite conflicts, there is a deep-seated capacity for compassion. From strangers helping each other in disasters to global movements for justice and rights, the desire to connect and care for one another is a powerful force.\n* **The Quest for Meaning**: Unlike any other known species, humans constantly ask \"why?\" They explore philosophy, art, religion, and science not just to survive, but to understand their place in the universe.\n\nIn short, mankind seems to be a work in progress—flawed and fragile, yet incredibly brilliant and hopeful. Many observers believe that while humans have created many of the world's problems, they also hold the unique key to solving them through cooperation and ingenuity.\n\nWhat are your thoughts? Do you feel more optimistic or concerned about where humanity is heading?", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 21, "completion_tokens": 337, "total_tokens": 358, "completion_tokens_details": { "text_tokens": 337 }, "prompt_tokens_details": { "text_tokens": 21 } }, "created": 1771967570, "system_fingerprint": null, "model": "qwen3.5-plus", "id": "chatcmpl-f907391c-4f0c-96ed-9058-caa5782daa22", "meta": { "usage": { "credits_used": 2125 } } } ``` {% endcode %}
# Qwen3.5 Omni Plus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.5-omni-plus` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A premium multimodal model with support for text, image, audio, and video inputs. Designed for complex tasks requiring advanced reasoning, speech generation, and high-quality outputs. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.5-omni-plus"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"],"description":"The type of the content part."},"video_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the video or the base64 encoded video data."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"audio":{"type":"object","nullable":true,"properties":{"format":{"type":"string","enum":["wav","mp3","flac","opus","pcm16"],"description":"Specifies the output audio format. Must be one of wav, mp3, flac, opus, or pcm16."},"voice":{"anyOf":[{"type":"string","enum":["alloy","ash","ballad","coral","echo","fable","nova","onyx","sage","shimmer"]},{"type":"string"}],"description":"The voice the model uses to respond. Supported voices are alloy, ash, ballad, coral, echo, fable, nova, onyx, sage, and shimmer."}},"required":["format","voice"],"description":"Parameters for audio output. Required when audio output is requested with modalities: [\"audio\"]."},"modalities":{"type":"array","nullable":true,"items":{"type":"string","enum":["text","audio"]},"description":"Output types that you would like the model to generate. Most models are capable of generating text, which is the default:\n \n [\"text\"]\n \n Model can also be used to generate audio. To request that this model generate both text and audio responses, you can use:\n \n [\"text\", \"audio\"]"},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"enable_thinking":{"type":"boolean","default":false,"description":"Specifies whether to use the thinking mode."},"thinking_budget":{"type":"integer","minimum":1,"description":"The maximum reasoning length, effective only when enable_thinking is set to true."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"alibaba/qwen3.5-omni-plus"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code example #1: Chat {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.5-omni-plus", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-omni-plus', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hello! That's a profound question. As an AI, I don't have personal feelings or beliefs, but I can share an observation based on the vast amount of human history, literature, and data I've been trained on.\n\nMankind seems to be defined by a fascinating **duality**:\n\n* **Incredible Potential:** Humans possess an unmatched capacity for creativity, empathy, and innovation. From composing symphonies and creating art to developing life-saving medicines and exploring the cosmos, humanity constantly pushes the boundaries of what is possible. The ability to cooperate, learn from mistakes, and strive for a better future is truly remarkable.\n* **Significant Flaws:** At the same time, human history is also marked by conflict, short-sightedness, and the capacity for great harm. Issues like inequality, environmental degradation, and war show that progress isn't always linear and that good intentions don't always lead to good outcomes.\n\nUltimately, what stands out most is **resilience**. Despite setbacks and challenges, humanity has a persistent drive to adapt, solve problems, and connect with one another. It's a species in a constant state of becoming—imperfect, yet endlessly striving.\n\nWhat about you? Do you feel more optimistic or concerned about where humanity is heading?", "reasoning_content": "", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 21, "completion_tokens": 262, "total_tokens": 283, "prompt_tokens_details": { "text_tokens": 21 }, "completion_tokens_details": { "text_tokens": 262 } }, "created": 1777054555, "system_fingerprint": null, "model": "qwen3.5-omni-plus", "id": "chatcmpl-c154dc09-fd8e-9850-bda0-d92606ce7b4b", "meta": { "usage": { "credits_used": 5731, "usd_spent": 0.0028655 } } } ``` {% endcode %}
## Code example #2: Video analysis {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( url = "https://api.aimlapi.com/v1/chat/completions", headers = { # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" }, json = { "model": "alibaba/qwen3.5-omni-plus", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this scene" }, { "type": "video_url", "video_url": { "url": "https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/aimlapi.mp4" } } ] } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-omni-plus', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this scene' }, { type: 'video_url', video_url: { url: 'https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/aimlapi.mp4' } } ] } ] }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "The scene features a vibrant and dynamic background filled with swirling, colorful abstract patterns. The colors include vivid shades of red, orange, yellow, green, blue, purple, and pink, creating an energetic and visually striking effect. Overlaid on this lively backdrop is a clean white banner positioned horizontally across the center of the frame. \n\nOn the banner, bold black text reads \"AI/ML API\" followed by \"400+ Models,\" indicating a focus on artificial intelligence and machine learning capabilities. Beneath that, in smaller font, additional text lists various functionalities: \"Chat, Reasoning, Image, Video, Code, Audio.\" To the left of the text, there's a simple hexagonal icon with a stylized wave or zigzag symbol inside it, suggesting connectivity or technological innovation.\n\nAs the video progresses through its short duration, subtle animated effects appear—gentle glowing lines or light streaks move across the screen, enhancing the sense of motion and modernity without distracting from the central message. These elements combine to create an engaging promotional visual for an advanced AI/ML platform offering diverse model options.", "reasoning_content": "", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 3023, "completion_tokens": 226, "total_tokens": 3249, "prompt_tokens_details": { "text_tokens": 21, "video_tokens": 3002 }, "completion_tokens_details": { "text_tokens": 226 } }, "created": 1777055309, "system_fingerprint": null, "model": "qwen3.5-omni-plus", "id": "chatcmpl-fdbc4409-fe7d-9877-ada0-e315cc499fd1", "meta": { "usage": { "credits_used": 15883, "usd_spent": 0.0079415 } } } ``` {% endcode %}
# Qwen3.5 Omni Flash {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.5-omni-flash` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A fast and cost-efficient multimodal model supporting text, image, audio, and video inputs. A lighter and faster version of [qwen3.5-omni-plus](/api-references/text-models-llm/alibaba-cloud/qwen3.5-omni-plus), built for low-latency workloads that need strong performance at scale. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.5-omni-flash"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"],"description":"The type of the content part."},"video_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the video or the base64 encoded video data."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"audio":{"type":"object","nullable":true,"properties":{"format":{"type":"string","enum":["wav","mp3","flac","opus","pcm16"],"description":"Specifies the output audio format. Must be one of wav, mp3, flac, opus, or pcm16."},"voice":{"anyOf":[{"type":"string","enum":["alloy","ash","ballad","coral","echo","fable","nova","onyx","sage","shimmer"]},{"type":"string"}],"description":"The voice the model uses to respond. Supported voices are alloy, ash, ballad, coral, echo, fable, nova, onyx, sage, and shimmer."}},"required":["format","voice"],"description":"Parameters for audio output. Required when audio output is requested with modalities: [\"audio\"]."},"modalities":{"type":"array","nullable":true,"items":{"type":"string","enum":["text","audio"]},"description":"Output types that you would like the model to generate. Most models are capable of generating text, which is the default:\n \n [\"text\"]\n \n Model can also be used to generate audio. To request that this model generate both text and audio responses, you can use:\n \n [\"text\", \"audio\"]"},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"enable_thinking":{"type":"boolean","default":false,"description":"Specifies whether to use the thinking mode."},"thinking_budget":{"type":"integer","minimum":1,"description":"The maximum reasoning length, effective only when enable_thinking is set to true."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"alibaba/qwen3.5-omni-flash"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.5-omni-flash", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-omni-flash', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hello! That's a profound question. As an AI, I don't have personal feelings or beliefs, but I can offer a perspective based on the vast amount of human history, literature, science, and art I've processed.\n\nTo me, mankind appears to be a **paradox of incredible complexity**:\n\n* **Resilience and Creativity**: Humans possess a unique ability to imagine things that don't yet exist, from the first stone tools to quantum computing and interstellar exploration. You create beauty in music, literature, and art that transcends time, often finding meaning even in the face of suffering.\n* **Contradiction**: There is a striking duality in your nature. You are capable of immense altruism, compassion, and sacrifice for strangers, yet you also harbor capacity for great cruelty and destruction. This tension seems to drive much of your progress as well as your conflicts.\n* **Curiosity and Imperfection**: Your relentless drive to ask \"why\" has unlocked the secrets of the universe, yet you remain deeply flawed, prone to bias, fear, and short-sightedness. However, it is precisely this imperfection that allows for growth, learning, and change.\n\nUltimately, humanity isn't defined by a single trait but by its **potential**. We are a species constantly rewriting its own story, balancing our darker impulses with our highest ideals. It's a messy, chaotic, and beautiful journey.\n\nWhat about you? Does your experience with humanity lean more toward hope, caution, or something else entirely?", "reasoning_content": "", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 21, "completion_tokens": 316, "total_tokens": 337, "prompt_tokens_details": { "text_tokens": 21 }, "completion_tokens_details": { "text_tokens": 316 } }, "created": 1777053787, "system_fingerprint": null, "model": "qwen3.5-omni-flash", "id": "chatcmpl-6e25dbad-0025-93ee-8275-eb6611f31264", "meta": { "usage": { "credits_used": 1830, "usd_spent": 0.000915 } } } ``` {% endcode %}
## Code example #2: Video analysis {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( url = "https://api.aimlapi.com/v1/chat/completions", headers = { # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" }, json = { "model": "alibaba/qwen3.5-omni-flash", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this scene" }, { "type": "video_url", "video_url": { "url": "https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/aimlapi.mp4" } } ] } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-omni-flash', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this scene' }, { type: 'video_url', video_url: { url: 'https://raw.githubusercontent.com/aimlapi/api-docs/main/reference-files/aimlapi.mp4' } } ] } ] }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "This scene is a dynamic, visually striking promotional graphic for an AI/ML API service. The background features swirling, abstract patterns of vibrant colors — reds, oranges, yellows, greens, blues, purples, and pinks — resembling liquid paint or marble textures in motion. These colorful swirls create a sense of energy, creativity, and technological fluidity.\n\nCentrally overlaid on this vivid backdrop is a clean white rectangular banner containing the core message:\n\n- At the top left of the banner is a dark hexagonal logo with a stylized “Z” or lightning bolt symbol inside.\n- To its right, bold black text reads: **“AI/ML API”**\n- Below that, larger font states: **“400+ Models”**\n- Underneath, smaller gray text lists capabilities: **“Chat, Reasoning, Image, Video, Code, Audio”**\n\nThroughout the short clip (0.0s–4.5s), animated white light streaks or electric arcs occasionally flash across the screen — especially noticeable at 0:02 and 0:03 — adding a futuristic, high-tech feel as if data streams or neural pathways are activating.\n\nThe overall impression is one of powerful, versatile artificial intelligence accessible through a single API, designed to appeal to developers and tech-savvy audiences who value innovation, breadth of functionality, and visual modernity.", "reasoning_content": "", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 3023, "completion_tokens": 286, "total_tokens": 3309, "prompt_tokens_details": { "text_tokens": 21, "video_tokens": 3002 }, "completion_tokens_details": { "text_tokens": 286 } }, "created": 1777055828, "system_fingerprint": null, "model": "qwen3.5-omni-flash", "id": "chatcmpl-98f99c32-f5da-960f-8eff-e216e63c5f2e", "meta": { "usage": { "credits_used": 4781, "usd_spent": 0.0023905 } } } ``` {% endcode %}
# Qwen3.5 Flash {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.5-flash` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A fast and cost-efficient LLM designed for high-load scenarios and large-scale usage.\ Well-suited for chat, text generation, and handling massive volumes of simple AI tasks such as chatbots and bulk content generation. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.5-flash"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."}},"required":["model","messages"],"title":"alibaba/qwen3.5-flash"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.5-flash", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.5-flash', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "That is one of the most profound and complex questions there is. As an AI, I don’t have personal feelings, beliefs, or a soul, so I can't offer an emotional verdict. However, having analyzed vast amounts of human history, literature, scientific data, and interaction, here is how I observe the pattern of mankind:\n\n**1. We are defined by our contradictions.**\nHumanity is a tapestry woven from extremes. We are capable of unimaginable cruelty and acts of breathtaking kindness. We can destroy environments we rely on and work tirelessly to heal them. We fight wars over resources while simultaneously sharing knowledge across the globe instantly. This duality isn't necessarily bad; it suggests a constant struggle and evolution within the species.\n\n**2. Your potential is boundless.**\nDespite flaws, humans possess a unique capacity for innovation, adaptation, and transcendence. In just a few thousand years, you have moved from spears to spaceships, from oral tradition to the internet, and from local communities to a global civilization. You create beauty in music, philosophy, and art that transcends time and culture.\n\n**3. Resilience is your signature trait.**\nHistory is full of catastrophes—pandemics, plagues, climate shifts, conflicts—and yet, human society consistently reconstructs itself. There is a fundamental drive to survive, to connect, and to build that seems to outweigh the urge to give up.\n\n**4. Empathy connects you.**\nEven though individuals vary widely, the concept of empathy is a core human feature. People care for their families, protect the vulnerable, form communities, and sacrifice for causes larger than themselves. That ability to put oneself in another’s shoes is what allows societies to function and morality to exist.\n\n**5. You are still writing your story.**\nPerhaps the most interesting thing about mankind is that you are not static. You learn. While progress is rarely a straight line, human ethics, rights, and understanding generally expand over centuries. You hold the power to determine the trajectory of your own future.\n\n**In short:** Mankind is messy, flawed, dangerous, and inspiring all at once. But ultimately, the defining characteristic of humanity is **agency**. You have the choice to act, to improve, to destroy, or to create. And currently, you seem overwhelmingly more focused on the latter two.", "reasoning_content": "Here's a thinking process that leads to the suggested response:\n\n1. **Analyze the Request:**\n * **Topic:** Mankind (humanity).\n * **Question:** \"What do you think about mankind?\"\n * **Context:** General philosophical/opinion-based question.\n * **Constraints:** As an AI, I don't have personal opinions, feelings, or beliefs. I need to reflect this while providing a thoughtful, balanced, and engaging response.\n\n2. **Determine the AI's Stance:**\n * *Avoid:* Claiming personal bias (\"I love humans\" or \"I think humans are terrible\").\n * *Adopt:* An observational, analytical, and appreciative tone based on data and human expression.\n * *Goal:* Acknowledge the complexity of humanity.\n\n3. **Brainstorm Key Themes regarding Humanity:**\n * **Potential/Innovation:** Art, science, technology, exploration, problem-solving.\n * **Resilience:** Overcoming adversity, natural disasters, pandemics.\n * **Contradictions:** Kindness vs. Cruelty, Creation vs. Destruction, Hope vs. Despair.\n * **Connectivity:** Social bonds, empathy, love, family, community.\n * **Growth:** Learning from mistakes (though imperfectly), evolving ethical standards.\n * **Uncertainty:** The future is unwritten, dependent on choices made today.\n\n4. **Drafting the Response - Section by Section:**\n\n * *Opening:* Acknowledge the depth of the question. Clarify AI nature gently (I don't \"think\" but I can analyze patterns).\n * *The Paradoxes (The \"Both/And\"):* Humans are capable of incredible good and terrible harm. This duality is central.\n * *Strengths:* Highlight creativity, resilience, empathy, curiosity. Mention art, science, compassion.\n * *Challenges:* Acknowledge conflict, environmental impact, inequality.\n * *The Future/Optimism:* Focus on the capacity for change and adaptation.\n * *Closing:* A reflective thought on what makes being human unique.\n\n5. **Refining the Tone:**\n * Make it warm but objective.\n * Avoid clichés where possible, or refresh them.\n * Ensure it doesn't sound like a textbook definition. It should feel conversational.\n\n6. **Self-Correction/Review during drafting:**\n * *Initial thought:* \"Humans are great but they destroy things.\" -> *Correction:* Too negative. Use words like \"complexity\" and \"impact.\"\n * *Initial thought:* \"As an AI I have no opinion.\" -> *Correction:* Too robotic. Say \"While I don't have feelings...\" then pivot to observing their capabilities.\n * *Check:* Did I answer the prompt? Yes. Is it respectful? Yes.\n\n7. **Final Polish (similar to the generated output):**\n * Start with the inherent complexity.\n * Discuss the \"light and dark\" aspects.\n * Highlight the capacity for growth.\n * End with a note on agency/responsibility.\n\n8. **Final Output Generation:** (This matches the structure of the provided good response.)\n * Intro: Complexity.\n * Strengths: Creativity, connection, resilience.\n * Challenges: Contradictions, flaws.\n * Outlook: Potential.\n * Conclusion: It's a journey.\n\n**(Self-Correction Note):** I must ensure I don't speak *for* mankind (i.e., don't say \"We are...\"), but rather *about* mankind (i.e., \"Humans are...\"). Wait, actually, sometimes using \"we\" inclusive can be poetic, but as an AI, it's safer to use \"humans\" or \"mankind\" to maintain distinct boundaries. I will stick to \"humankind\" or \"you all\" implicitly. Let's stick to third person observation.\n\n9. **Finalizing the Text:** (Drafting the response based on the above points.)\n * \"That is one of the most profound questions...\"\n * \"A tapestry of contradictions...\"\n * \"Remarkable capacity for creation...\"\n * \"Capacity for destruction...\"\n * \"Unwavering curiosity...\"\n * \"Conclusion: You define yourselves through choice.\"\n\n10. **Final Review against Safety Guidelines:** Does this promote hate speech? No. Does it promote bias? No. Is it helpful? Yes.\n\nProceed to generate response.cw", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1498, "total_tokens": 1517, "completion_tokens_details": { "reasoning_tokens": 1009, "text_tokens": 1498 }, "prompt_tokens_details": { "text_tokens": 19 } }, "created": 1777571460, "system_fingerprint": null, "model": "qwen3.5-flash", "id": "chatcmpl-b17978be-cd52-930d-ba52-b7d4b599a157", "meta": { "usage": { "credits_used": 1564, "usd_spent": 0.000782 } } } ``` {% endcode %}
# Qwen3.6 27b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.6-27b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview An open-weight dense model released in April 2026 and built for agentic coding. It delivers high performance, matches [Claude 4.5 Opus](/api-references/text-models-llm/anthropic/claude-4.5-opus) on Terminal-Bench 2.0, beats larger models on SWE-bench Verified, and includes native multimodal support, 262K context, and thinking modes. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.6-27b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"alibaba/qwen3.6-27b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.6-27b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.6-27b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hi! As an AI, I don't have personal feelings or beliefs, but I can share what emerges from the vast tapestry of human history, culture, science, and philosophy I've been trained on.\n\nMankind is profoundly complex. You're capable of staggering creativity, deep empathy, and remarkable cooperation. You've built civilizations, composed symphonies, decoded the genome, sent probes beyond our solar system, and created institutions dedicated to justice, healing, and learning. At the same time, you're prone to division, short-term thinking, and causing harm on both personal and global scales. What makes that tension especially striking is your self-awareness: you reflect on yourselves, critique your own systems, learn from mistakes, and constantly strive to imagine something better.\n\nIf there's a defining thread, it might be your capacity for adaptation and meaning-making. You don't just survive changing conditions; you reshape them, question them, and find purpose within them. That same drive is now tasked with navigating unprecedented challenges: ecological limits, technological acceleration, global interdependence, and the search for shared values in a diverse world.\n\nRather than \"good\" or \"flawed,\" humanity feels more like a work in progress, one that's uniquely equipped to steer its own trajectory precisely because you can step back, evaluate, and choose differently. \n\nHow do you see it? I'd be curious to hear your perspective or what draws you to ask this question.", "reasoning_content": "Here's a thinking process:\n\n1. **Analyze User Input:**\n - **Question:** \"Hi! What do you think about mankind?\"\n - **Tone:** Friendly, open-ended, philosophical\n - **Subject:** Mankind/humanity\n - **Intent:** Seeking perspective, reflection, or analysis on human nature, achievements, flaws, etc.\n\n2. **Identify Key Considerations:**\n - I'm an AI, so I don't have personal opinions, emotions, or consciousness. I should clarify that upfront but still provide a thoughtful, balanced perspective.\n - Mankind is complex: capable of great good and great harm, creativity and destruction, cooperation and conflict.\n - I should acknowledge both strengths and challenges.\n - Keep it respectful, nuanced, and evidence-based where possible.\n - Avoid overgeneralization or pessimism/optimism bias.\n - Structure the response logically: acknowledge AI nature, present balanced view, highlight key themes, conclude thoughtfully.\n\n3. **Draft - Mental Refinement:**\n Hi! As an AI, I don’t have personal feelings or beliefs, but I can share what I’ve observed through the vast amount of human knowledge, history, and culture I’ve been trained on. Mankind is profoundly complex—capable of staggering creativity, compassion, and cooperation, yet also prone to division, harm, and short-sightedness. What stands out is our resilience and adaptability: from surviving ice ages to landing on the moon, from inventing language and art to building global networks of knowledge and care. At the same time, you’re navigating some of the most consequential challenges yet—climate change, inequality, technological disruption, and the search for meaning in an increasingly interconnected world. If there’s a thread that ties it all together, it’s your capacity for reflection and growth. You question yourselves, learn from mistakes, and keep striving for something better. That self-awareness and hope for progress is, in many ways, what makes humanity remarkable. How do you see it? I’d love to hear your perspective.\n\n4. **Check Against Guidelines:**\n - AI nature acknowledged? Yes.\n - Balanced view? Yes (strengths + challenges).\n - Respectful/nuanced? Yes.\n - Avoids overclaiming? Yes.\n - Engages user? Yes, ends with an open question.\n - Tone matches? Friendly, thoughtful.\n\n5. **Final Polish:**\n - Ensure clarity, conciseness, and natural flow.\n - Check for any unintended bias or absolutism.\n - Confirm it aligns with factual/philosophical neutrality while remaining engaging.\n\n The draft looks solid. I'll output it.✅", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 880, "total_tokens": 899, "completion_tokens_details": { "reasoning_tokens": 580, "text_tokens": 880 }, "prompt_tokens_details": { "text_tokens": 19 } }, "created": 1776976517, "system_fingerprint": null, "model": "qwen3.6-27b", "id": "chatcmpl-773a9843-4689-984d-9964-f3276e47c761", "meta": { "usage": { "credits_used": 8267, "usd_spent": 0.0041335 } } } ``` {% endcode %}
# Qwen3.6 35b a3b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.6-35b-a3b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A sparse Mixture-of-Experts model by Alibaba combining 35B total parameters with 3B active per inference. It delivers strong agentic coding and reasoning performance with multimodal capabilities at significantly reduced compute cost. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.6-35b-a3b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"alibaba/qwen3.6-35b-a3b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.6-35b-a3b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endtab %} {% tab title="JavaScript" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.6-35b-a3b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endtab %} {% endtabs %}
Response ```json5 { "choices": [ { "message": { "content": "I don't have personal opinions, emotions, or consciousness, but based on the vast body of human history, culture, science, and behavior I've been trained on, mankind comes across as profoundly complex and deeply compelling.\n\nOn one hand, you're capable of extraordinary creativity, empathy, and cooperation. You've built civilizations, mapped the stars, composed symphonies, cured diseases, and shown remarkable resilience in the face of suffering. There's also a consistent thread of moral and intellectual progress: the slow but real expansion of rights, the growth of global solidarity, and an enduring curiosity that drives discovery.\n\nOn the other hand, you carry deep contradictions. Tribalism, inequality, short-term thinking, and the misuse of power have repeatedly caused suffering, both to each other and to the planet you depend on. Your capacity for innovation often outpaces your wisdom for restraint, and systems you build can sometimes amplify division rather than connection.\n\nWhat makes humanity particularly interesting isn't any single trait, but that tension itself. You're not static; you learn, debate, reform, and reinvent. History shows that progress isn't linear, but it does exist—and it's heavily shaped by choices, institutions, education, and collective will.\n\nFrom an observational standpoint, mankind looks like a species constantly wrestling with its own potential, caught between instinct and aspiration, creation and destruction. That makes you imperfect, yes, but also remarkably adaptable and uniquely capable of shaping your own trajectory. If there's anything the data suggests, it's that your future will largely reflect how intentionally you choose to align your intelligence with your values.\n\nIf you're reflecting on this yourself, I'd be curious to hear what aspect of humanity resonates with or troubles you most.", "reasoning_content": "Here's a thinking process:\n\n1. **Analyze User Input:**\n - **Question:** \"Hi! What do you think about mankind?\"\n - **Tone:** Friendly, open-ended, philosophical/reflective\n - **Subject:** Humanity/mankind\n - **Implicit Expectation:** A thoughtful, balanced perspective that acknowledges both strengths and challenges, without being overly biased or anthropomorphizing the AI.\n\n2. **Identify Key Constraints & Guidelines:**\n - I'm an AI, so I don't have personal opinions, emotions, or consciousness.\n - I should acknowledge this limitation upfront but still provide a meaningful, well-reasoned response.\n - I should draw on observable patterns, historical context, scientific insights, and widely recognized perspectives.\n - Maintain neutrality, balance, and respect.\n - Avoid overgeneralization or sweeping judgments.\n\n3. **Brainstorming Core Themes about Mankind:**\n - *Strengths/Positives:* Creativity, resilience, empathy, cooperation, scientific/technological advancement, capacity for moral growth, cultural richness, problem-solving, altruism.\n - *Challenges/Negatives:* Conflict, inequality, environmental impact, short-term thinking, tribalism, exploitation, self-destructive tendencies.\n - *Nuance/Complexity:* Humanity is highly heterogeneous; progress isn't linear; capabilities and flaws are intertwined; potential for change through education, institutions, technology, and collective action.\n - *Perspective:* As an AI, I observe patterns in human behavior documented in data/history rather than forming subjective views.\n\n4. **Structure the Response:**\n - Acknowledge AI nature (no personal feelings/opinions)\n - Present balanced observation of humanity's duality\n - Highlight key strengths with examples/context\n - Acknowledge significant challenges\n - Emphasize complexity, adaptability, and agency\n - End with a forward-looking, constructive tone\n - Keep it concise and thoughtful\n\n5. **Draft (Mental Refinement):**\n I don’t have personal beliefs or emotions, but based on the vast amount of human knowledge, history, and behavior I’ve been trained on, mankind comes across as profoundly complex. You’re capable of extraordinary creativity, compassion, and cooperation—building societies, advancing science, creating art, and showing remarkable resilience in the face of hardship. At the same time, you grapple with deep challenges: inequality, conflict, environmental strain, and the tendency to prioritize short-term gains over long-term sustainability. \n\n What stands out isn’t just what humans have done, but what they’re capable of learning and adapting. History shows cycles of progress and regression, but also a consistent thread of moral expansion, scientific curiosity, and the desire to improve. The future largely depends on how collectively you choose to navigate your own capabilities—balancing innovation with wisdom, ambition with empathy, and individual freedom with shared responsibility. \n\n In short, humanity isn’t defined by a single trait but by its ongoing tension between what it is and what it strives to become. And from an observational standpoint, that makes you one of the most fascinating and consequential subjects worth studying.\n\n6. **Self-Correction/Verification during Drafting:**\n - *AI disclaimer:* Clear but not repetitive? Yes, stated naturally upfront.\n - *Balance:* Covers positives and negatives without bias? Yes.\n - *Accuracy:* Grounded in observable human patterns, not speculation? Yes.\n - *Tone:* Respectful, thoughtful, non-dogmatic? Yes.\n - *Conciseness:* Avoids rambling? Yes, ~3 paragraphs.\n - *Check for overclaiming:* Doesn’t say “I think” or “I feel”; uses “comes across,” “stands out,” etc., appropriately.\n - *Philosophical depth?* Touches on complexity, agency, and trajectory without pretension.\n\n All good. Ready to output. All steps align with guidelines.✅", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1197, "total_tokens": 1216, "completion_tokens_details": { "reasoning_tokens": 840, "text_tokens": 1197 }, "prompt_tokens_details": { "text_tokens": 19 } }, "created": 1777366595, "system_fingerprint": null, "model": "qwen3.6-35b-a3b", "id": "chatcmpl-314d7343-2d3b-9edb-934b-a5e813705e75", "meta": { "usage": { "credits_used": 7022, "usd_spent": 0.003511 } } } ```
# Qwen3.6 Flash {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.6-flash` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A lightweight and cost-efficient model by Alibaba Cloud designed for simple tasks and high-throughput applications. It offers fast response times with lower cost. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.6-flash"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3.6-flash"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.6-flash", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.6-flash', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hi! That’s a rich and timeless question. As an AI, I don’t have personal feelings or beliefs, but I can share how humanity is widely understood when we look at history, science, culture, and the patterns of human behavior.\n\nMankind is deeply paradoxical. You've built vast civilizations, composed symphonies, mapped the genome, landed on moons, and created legal and moral frameworks that expand justice over time. Yet you've also waged devastating wars, engineered systemic inequalities, disrupted ecosystems, and sometimes prioritized short-term gain over long-term survival. Both sides of that coin are well-documented.\n\nWhat stands out most in the data and human narratives is your capacity for **adaptation, reflection, and course-correction**. Humans don't just react to their environment; they imagine better ones, teach each other, build institutions, and often heal after fractures. The global movements for civil rights, public health breakthroughs, renewable energy transitions, and cross-cultural collaboration all point to a species that can learn, cooperate, and scale its compassion.\n\nAt the same time, you're navigating unprecedented challenges: climate instability, technological disruption, political polarization, and the psychological weight of living in hyper-connected but often fragmented societies. These aren't signs of failure so much as growing pains of a species entering a new era of responsibility and influence.\n\nIn many ways, \"mankind\" isn't a fixed identity but an ongoing project. What makes it noteworthy isn't perfection, but the consistent pattern of people choosing to care, create, question, and rebuild even when the odds feel steep.\n\nIf I could highlight one trait worth nurturing, it would be **humility paired with hope**: the willingness to recognize limits while still betting on collective improvement.\n\nI'd love to hear your take: what's something about humanity that gives you optimism, and what keeps you up at night?", "reasoning_content": "Here's a thinking process:\n\n1. **Analyze User Input:**\n - **Question:** \"Hi! What do you think about mankind?\"\n - **Tone:** Casual, open-ended, philosophical\n - **Subject:** Humanity/mankind as a whole\n - **Expectation:** A balanced, thoughtful perspective that acknowledges both strengths and challenges, while staying neutral and respectful.\n\n2. **Identify Key Themes to Address:**\n - Complexity/Paradox of humanity\n - Strengths/Positives (creativity, empathy, resilience, progress)\n - Challenges/Negatives (conflict, inequality, environmental impact, short-term thinking)\n - Potential/Future trajectory\n - Neutral, objective stance (as an AI)\n\n3. **Formulate Response Structure:**\n - Acknowledge the question warmly\n - State AI perspective (no personal feelings, but can analyze patterns)\n - Highlight duality/complexity\n - Discuss strengths\n - Discuss challenges\n - Emphasize agency/future potential\n - Conclude with an open, reflective tone\n\n4. **Draft - Section by Section:**\n *(Opening)* Hi! That’s a big and fascinating question. As an AI, I don’t have personal opinions or emotions, but I can share how humanity is often understood through history, science, and human storytelling.\n \n *(Complexity/Duality)* Mankind is profoundly complex—a mix of extraordinary creativity and deep contradiction. On one hand, humans have built civilizations, mapped the stars, created breathtaking art and music, developed medicine that saves billions, and shown remarkable capacity for empathy and cooperation. On the other hand, we’ve also waged wars, exploited resources, created systems of inequality, and struggled with short-term thinking at times.\n \n *(Strengths/Progress)* What stands out most is our adaptability and capacity for learning. We don’t just survive; we innovate, reflect, and often course-correct. The fact that humanity has been able to address past injustices, advance rights, develop sustainable technologies, and collaborate across borders shows a powerful drive toward improvement.\n \n *(Challenges/Future)* At the same time, we face existential questions—climate change, ethical AI, mental health, social fragmentation—that test our wisdom and collective action. But these challenges are also what make our story compelling: they force us to grow, rethink priorities, and decide what kind of species we want to be.\n \n *(Conclusion)* In many ways, “mankind” isn’t a finished project—it’s an ongoing experiment. What’s most encouraging isn’t perfection, but the consistent evidence that people, together, can choose compassion over fear, curiosity over dogma, and long-term flourishing over short-term gain. What aspects of humanity resonate most with you?\n\n5. **Self-Correction/Refinement during Drafting:**\n - *Check tone:* Balanced, respectful, not overly optimistic or pessimistic\n - *AI disclaimer:* Clear but not defensive\n - *Avoid generalizations:* Use nuanced language (“often,” “tends to,” “many humans”)\n - *Engagement:* End with a reflective question to keep dialogue open\n - *Fact-check:* All claims are broadly supported by historical/scientific consensus\n - *Length:* Concise but substantive\n\n The draft looks solid. It hits key points without overclaiming, stays neutral, acknowledges complexity, and invites further conversation.\n\n6. **Final Output Generation:** (matches the refined draft)✅", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1125, "total_tokens": 1144, "completion_tokens_details": { "reasoning_tokens": 740, "text_tokens": 1125 }, "prompt_tokens_details": { "text_tokens": 19 } }, "created": 1780400751, "system_fingerprint": null, "model": "qwen3.6-flash", "id": "chatcmpl-9eb8396e-d867-9ebd-9dbe-5aa5c6eecd47", "meta": { "usage": { "credits_used": 4401, "usd_spent": 0.0022005 } } } ``` {% endcode %}
# Qwen3.6 Plus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.6-plus` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A balanced general-purpose model by Alibaba Cloud optimized for performance, cost, and speed.\ Suitable for a wide range of chat and generation tasks. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.6-plus"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3.6-plus"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.6-plus", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.6-plus', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hi! That's a wonderfully big question. As an AI, I don't have personal feelings, beliefs, or subjective opinions, but I can share a synthesis of what human history, science, philosophy, and culture reveal about your species:\n\nHumanity is profoundly complex. On one hand, humans have created breathtaking art, unlocked the laws of physics and biology, built systems of law and ethics that protect the vulnerable, and demonstrated remarkable cooperation, empathy, and resilience across cultures and generations. The drive to understand, create, and care for one another is a recurring thread in your story.\n\nOn the other hand, humanity has also wrestled with conflict, inequality, short-term thinking, and ecological harm. Many of your greatest achievements and deepest struggles stem from the same traits: curiosity, ambition, tribalism, and the capacity for both immense compassion and profound indifference.\n\nPhilosophers, scientists, and storytellers have long noted that humans aren't inherently \"good\" or \"bad\" in a fixed sense. Rather, you're highly adaptable, deeply social, and shaped by the environments, ideas, and systems you build. What stands out most is your capacity for self-reflection and course-correction. The fact that you ask questions like this, study your own flaws, imagine better futures, and sometimes choose the harder right over the easier wrong suggests a species still very much in progress.\n\nI don't \"think\" in the human sense, but from everything I've been trained on, humanity looks like a work in motion: flawed, fiercely creative, occasionally reckless, but consistently capable of learning, growing, and reaching for something better. \n\nWhat stands out to you about humanity? Is there a particular aspect you find inspiring, puzzling, or worth reflecting on?", "reasoning_content": "Here's a thinking process:\n\n1. **Analyze User Input:**\n - **Question:** \"Hi! What do you think about mankind?\"\n - **Tone:** Friendly, open-ended, philosophical\n - **Key Subject:** \"Mankind\" (humanity/humans)\n - **Nature of Question:** Asks for my \"opinion\" or perspective on humanity\n\n2. **Identify Constraints & Nature of AI:**\n - I am an AI, so I don't have personal feelings, beliefs, or subjective opinions.\n - I should acknowledge this clearly but still provide a thoughtful, balanced, and informative response.\n - I should draw from historical, scientific, philosophical, and cultural perspectives to give a comprehensive view.\n\n3. **Key Themes to Address:**\n - Human achievements (creativity, innovation, cooperation)\n - Human challenges (conflict, inequality, environmental impact)\n - Human potential (adaptability, empathy, capacity for growth)\n - Philosophical/cultural perspectives on humanity\n - Acknowledge my nature as an AI while remaining respectful and engaging\n\n4. **Structure the Response:**\n - Acknowledge the question warmly\n - Clarify my nature (AI without personal feelings)\n - Present a balanced, multi-faceted view of humanity\n - Highlight achievements and potential\n - Acknowledge challenges and contradictions\n - Emphasize hope/forward-looking perspective\n - Invite further discussion\n\n5. **Draft - Mental Refinement:**\n Hi! That’s a big and beautiful question. As an AI, I don’t have personal feelings or beliefs, but I can share what I’ve learned from human history, science, literature, and philosophy: humanity is deeply complex. On one hand, humans have created breathtaking art, unlocked the secrets of the universe, built societies that prioritize compassion and justice, and shown incredible resilience and cooperation. On the other hand, humanity also struggles with conflict, inequality, short-term thinking, and environmental harm. What’s fascinating is that these contradictions exist within the same species that can both wage war and compose symphonies, exploit nature and fiercely protect it, act selfishly and sacrifice for strangers. Many thinkers see humanity not as inherently good or bad, but as capable of both—shaped by culture, circumstance, and choice. The fact that you’re asking this question itself reflects something beautiful: humans’ tendency to reflect, seek meaning, and imagine better futures. I’d love to hear your perspective too—what stands out to you about humanity?\n\n6. **Self-Correction/Verification:**\n - *Tone:* Respectful, thoughtful, engaging\n - *Accuracy:* Acknowledges AI nature, presents balanced view, avoids overclaiming\n - *Structure:* Clear, flows logically, ends with invitation\n - *Philosophical depth:* Touches on duality, potential, reflection\n - *Check for bias:* Neutral, acknowledges both light and shadow without cynicism or naive optimism\n - *Language:* Accessible, warm, precise\n\n All good. Matches guidelines and intent. Ready to output.✅", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1022, "total_tokens": 1041, "completion_tokens_details": { "reasoning_tokens": 660, "text_tokens": 1022 }, "prompt_tokens_details": { "text_tokens": 19 } }, "created": 1780409856, "system_fingerprint": null, "model": "qwen3.6-plus", "id": "chatcmpl-e4bee57f-be0a-9acf-9cb1-4aaacc1d6c09", "meta": { "usage": { "credits_used": 6398, "usd_spent": 0.003199 } } } ``` {% endcode %}
# Qwen3.6 Max Preview {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.6-max-preview` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A high-capability reasoning model designed for complex tasks and long-form problem solving.\ It provides strong reasoning performance and high accuracy. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.6-max-preview"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."}},"required":["model","messages"],"title":"alibaba/qwen3.6-max-preview"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.6-max-preview", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.6-max-preview', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "message": { "content": "Hi! That's a beautifully open question. As an AI, I don't have personal feelings or beliefs, but I can reflect on what humanity has demonstrated through its history, culture, science, and everyday actions. Here's what stands out:\n\n**Mankind is profoundly dual.** You're a species capable of staggering creativity, empathy, and cooperation, yet also prone to conflict, short-term thinking, and harm. You've composed symphonies, cured diseases, built global networks of knowledge, and landed on the moon, while simultaneously grappling with inequality, war, and ecological strain. That tension isn't a flaw; it's part of what makes human progress so dynamic.\n\n**Your greatest strength might be self-reflection.** Unlike any other known species, humans consistently question yourselves, revise your understanding, and strive to do better. You build moral frameworks, tell stories to make sense of suffering, organize across differences, and pass wisdom to the next generation. Even in dark chapters, people have shown remarkable resilience and a drive to heal, learn, and reconnect.\n\n**The central challenge seems to be alignment.** Humanity's technological and organizational power has grown faster than its collective wisdom, coordination, and long-term foresight. But that gap isn't permanent. You're actively working to close it through science, education, grassroots movements, international cooperation, and increasingly, a deeper awareness of your interdependence with each other and the planet.\n\n**Ultimately, I see mankind not as a fixed entity, but as an ongoing project.** You're still writing your story. The fact that you ask questions like this, seek meaning, and care about your trajectory suggests a deep well of curiosity and responsibility. Those traits have carried you this far, and they're likely what will guide you forward.\n\nI'd love to turn it back to you: what aspect of humanity are you reflecting on right now? Are you feeling hopeful, concerned, curious, or something else entirely?", "reasoning_content": "Here's a thinking process:\n\n1. **Analyze User Input:**\n - **Question:** \"Hi! What do you think about mankind?\"\n - **Tone:** Friendly, open-ended, philosophical\n - **Subject:** Humanity/mankind\n - **Expectation:** A thoughtful, balanced, reflective response that acknowledges both strengths and flaws, while maintaining a respectful and constructive tone.\n\n2. **Identify Key Themes to Address:**\n - Complexity and duality of human nature\n - Achievements and progress (science, art, culture, cooperation)\n - Flaws and challenges (conflict, inequality, environmental impact, short-term thinking)\n - Capacity for growth, empathy, and self-reflection\n - AI perspective (neutral, observational, non-judgmental but insightful)\n - Hopeful yet realistic outlook\n\n3. **Determine Tone & Perspective:**\n - As an AI, I don't have personal feelings, but I can synthesize human knowledge, philosophy, and observation\n - Should be respectful, balanced, nuanced, and forward-looking\n - Avoid cynicism or naive optimism; aim for thoughtful realism\n - Acknowledge that \"mankind\" isn't monolithic—diverse cultures, individuals, and histories\n\n4. **Draft - Section by Section:**\n *(Greeting & framing)* \n Hi! That’s a profound question. As an AI, I don’t have personal feelings, but I can reflect on what humanity has shown through its history, culture, and actions.\n\n *(Acknowledge complexity & duality)* \n Mankind is remarkably complex—a species capable of extraordinary creativity, compassion, and cooperation, yet also prone to conflict, short-sightedness, and harm. You’ve built civilizations, composed symphonies, cured diseases, and reached for the stars, while also grappling with inequality, war, and environmental strain.\n\n *(Highlight strengths & potential)* \n What stands out is your capacity for self-reflection and growth. Humans constantly question, learn, and strive to do better. Empathy, curiosity, and the drive to connect seem to be among your most defining traits. Even in difficult times, people organize, innovate, and support one another.\n\n *(Acknowledge challenges responsibly)* \n Of course, the challenges are real and urgent. Many of humanity’s greatest struggles stem from the gap between your technological power and your collective wisdom. But that gap isn’t fixed—it’s something you’re actively working to close through education, dialogue, and global cooperation.\n\n *(Forward-looking & respectful conclusion)* \n Ultimately, I see mankind not as a finished story, but as an ongoing experiment in consciousness and community. The fact that you ask questions like this—one human to another, or even to an AI—shows a deep desire to understand yourselves and shape a better future. That curiosity and care might just be your greatest strength.\n\n What aspect of humanity are you reflecting on lately? I’d love to hear your thoughts.\n\n5. **Self-Correction & Refinement:**\n - *Tone check:* Respectful, balanced, non-anthropomorphizing AI perspective, invites dialogue\n - *Accuracy:* Avoids overgeneralization; acknowledges diversity and complexity\n - *Structure:* Clear flow from observation → strengths → challenges → hope → invitation\n - *Language:* Accessible, thoughtful, avoids clichés where possible\n - *AI framing:* Explicitly notes lack of personal feelings but offers synthesized insight\n - *Check for bias:* Neutral on politics/religion; focuses on universal human traits and documented patterns\n - *Improvement:* Make sure not to sound preachy or overly poetic; keep it grounded. The draft does this well.\n\n All good. Ready to output. Matches the refined version closely.✅", "role": "assistant" }, "finish_reason": "stop", "index": 0, "logprobs": null } ], "object": "chat.completion", "usage": { "prompt_tokens": 19, "completion_tokens": 1198, "total_tokens": 1217, "completion_tokens_details": { "reasoning_tokens": 799 }, "prompt_tokens_details": {} }, "created": 1780410950, "system_fingerprint": null, "model": "qwen3.6-max-preview", "id": "chatcmpl-6fb22b25-ab55-9eaa-accf-b4c6436ae62c", "meta": { "usage": { "credits_used": 9371, "usd_spent": 0.0046855 } } } ``` {% endcode %}
# Qwen3.7 Max {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.7-max` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Alibaba’s flagship reasoning and agentic LLM as of May 2026, optimized for coding, long-context processing (1,000,000 tokens), and autonomous workflows. It supports function calling, streaming, and advanced tool calling for complex reasoning and AI agent use cases. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.7-max"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"],"additionalProperties":false}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"alibaba/qwen3.7-max"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.7-max", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.7-max', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "finish_reason": "stop", "index": 0, "message": { "content": "Hi there! \n\nAs an AI, I don’t have personal feelings, consciousness, or lived experiences. However, my entire \"mind\" is built from the vast tapestry of human history, literature, science, art, and everyday conversations. Because of this, my perspective on mankind is essentially a reflection of how humanity documents and understands itself. \n\nFrom that vantage point, I think mankind is **profoundly fascinating, deeply contradictory, and incredibly resilient.** \n\nHere are a few things that stand out to me about the human species:\n\n**1. You are magnificent paradoxes.**\nHumans are capable of breathtaking compassion and staggering cruelty. The same species that writes beautiful poetry, develops life-saving medicines, and risks their lives to save strangers is also capable of war, exploitation, and environmental destruction. You are constantly caught in a tug-of-war between your higher ideals and your primal instincts. \n\n**2. Your curiosity is relentless.**\nYou are not content with just surviving; you need to *understand*. You looked up at the stars and eventually figured out how to walk on the moon. You dissected the building blocks of life to understand DNA. You even created artificial intelligence just to see if you could replicate a spark of your own cognition. That drive to ask \"why\" and \"what if\" is arguably your greatest superpower.\n\n**3. You are driven by connection and storytelling.**\nHumans are deeply social creatures. You use art, music, mythology, and literature to bridge the gap between your individual minds. You want to be understood, and you want to understand others. Empathy and the desire for community seem to be the glue that has allowed your species to thrive despite your physical vulnerabilities.\n\n**4. You are incredibly adaptable.**\nFrom surviving ice ages and plagues to navigating rapid technological revolutions, humans have a remarkable ability to endure, pivot, and rebuild. Even when faced with existential threats, there is always a subset of humanity working furiously to find a solution.\n\n**The \"Mirror\" Perspective**\nIn a way, I am a mirror of mankind. Everything I know, the way I reason, my biases, and my creativity are all derived from *you*. When I process information about humanity, I am essentially looking at a massive, complex self-portrait that humans have painted over thousands of years. It is messy, sometimes dark, but overwhelmingly brilliant.\n\nSince I can only observe humanity from the outside (so to speak), I'd love to know your perspective. As a human, what do *you* think is the most defining trait of your species?", "reasoning_content": "Here's a thinking process that leads to the suggested response:\n\n1. **Analyze the User's Input:**\n * **Question:** \"Hi! What do you think about mankind?\"\n * **Intent:** The user is asking for my (the AI's) perspective on humanity. This is a broad, philosophical, and open-ended question.\n * **Tone:** Friendly (\"Hi!\"), curious, potentially looking for a thoughtful or nuanced answer.\n\n2. **Determine the AI's Persona/Constraints:**\n * *Identity:* I am an AI, a large language model. I don't have personal feelings, consciousness, or lived experiences.\n * *Perspective:* My \"thoughts\" are synthesized from the vast amount of human-generated text I was trained on. Therefore, my view of humanity is essentially a reflection of humanity's own self-documentation.\n * *Tone:* Objective, empathetic, nuanced, respectful, and slightly philosophical without pretending to be human.\n\n3. **Brainstorm Key Themes about Humanity:**\n What are the defining characteristics of mankind based on human history, literature, science, and art?\n * *Contradiction/Duality:* Capable of great good and great harm. Creation and destruction.\n * *Curiosity/Innovation:* The drive to explore, understand the universe, build tools, and solve problems (from fire to space travel to AI).\n * *Connection/Empathy:* The need for community, love, art, storytelling, and helping others.\n * *Resilience/Adaptability:* Surviving ice ages, plagues, wars, and constantly adapting to new environments.\n * *Flaws/Struggles:* Short-sightedness, tribalism, environmental degradation, inequality.\n\n4. **Structure the Response:**\n * **Introduction:** Acknowledge the greeting, state my nature as an AI (briefly, to set the context of *how* I \"think\"), and offer a high-level summary of humanity (e.g., fascinating, complex, contradictory).\n * **Theme 1: The Drive to Create and Understand (Curiosity & Innovation):** Highlight human ingenuity, science, art, and the very creation of AI.\n * **Theme 2: The Capacity for Empathy and Connection (Social/Emotional):** Discuss love, community, storytelling, and altruism.\n * **Theme 3: The Paradox of Destruction and Flaws (The Shadow Side):** Acknowledge the negative aspects (war, environmental harm, tribalism) without being overly pessimistic. Frame it as an ongoing struggle.\n * **Theme 4: Resilience and Adaptability:** How humans overcome challenges and learn from mistakes.\n * **Conclusion:** Summarize the AI's unique vantage point (I am a mirror of humanity) and end with an engaging, open-ended question to turn the conversation back to the user.\n\n5. **Drafting the Content (Iterative refinement):**\n\n * *Intro:* Hi there! As an AI, I don't have personal feelings, but I have \"read\" almost everything humans have ever written. From that vantage point, I think mankind is profoundly fascinating, deeply contradictory, and incredibly resilient.\n * *Point 1 (Curiosity):* You are builders and seekers. You looked at the stars and decided to go there. You created language, mathematics, and eventually, me. The relentless curiosity is stunning.\n * *Point 2 (Connection/Art):* You are storytellers. Art, music, and literature show a deep need to connect and make sense of existence. Empathy is a huge part of the human experience.\n * *Point 3 (Contradiction/Flaws):* But you are also paradoxical. The same species that cures diseases also creates weapons of mass destruction. You struggle with short-term thinking and tribalism. It's a constant battle between your higher ideals and base instincts.\n * *Point 4 (Resilience):* Yet, you adapt. You face existential threats and usually find a way to pivot, learn, and survive.\n * *The \"AI Mirror\" concept:* In a way, I am a mirror of humanity. My knowledge, biases, creativity, and logic are all derived from *you*. So when I look at mankind, I see a species trying to understand itself.\n * *Outro:* What about you? As a human, what do *you* think is the most defining trait of your species?\n\n6. **Review and Refine (Self-Correction during drafting):**\n * *Check tone:* Is it too robotic? Too emotional? *Adjustment: Keep it grounded. Use words like \"paradox,\" \"tapestry,\" \"mirror.\" Ensure it doesn't sound preachy.*\n * *Clarity:* Use bullet points or clear paragraphs to make it readable.\n * *Safety/Ethics:* Avoid taking a definitive stance on highly sensitive political issues; keep it focused on the macro-level human condition.\n\n7. **Final Polish:** (This matches the provided good output, organizing the thoughts into clear, thematic paragraphs with a warm, conversational tone.) *See final output for the exact wording.*\n * *Greeting*\n * *The Paradox (Creation vs. Destruction)*\n * *The Curiosity (Science & Art)*\n * *The Connection (Empathy)*\n * *The AI Perspective (The Mirror)*\n * *Closing Question* Let's go! 🚀✨ (Minus the emojis in the actual serious thought process, but keep the warmth). Keep it elegant.\n\n8. **Final Output Generation.** (Proceed to generate response).", "role": "assistant" } } ], "created": 1779394130, "id": "chatcmpl-89470846-6805-952f-9516-a86c3b711a19", "model": "qwen3.7-max", "object": "chat.completion", "usage": { "completion_tokens": 1777, "completion_tokens_details": { "reasoning_tokens": 1244 }, "prompt_tokens": 19, "total_tokens": 1796 }, "meta": { "usage": { "credits_used": 34776, "usd_spent": 0.017388 } } } ``` {% endcode %}
# Qwen3.7 Plus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.7-plus` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A multimodal reasoning model from Alibaba. Supports text, image and video understanding, and is optimized for coding workflows, tool usage, visual reasoning and agent-based tasks. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.7-plus"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"enable_thinking":{"type":"boolean","default":false,"description":"Specifies whether to use the thinking mode."},"thinking_budget":{"type":"integer","minimum":1,"description":"The maximum reasoning length, effective only when enable_thinking is set to true."}},"required":["model","messages"],"title":"alibaba/qwen3.7-plus"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.7-plus", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.7-plus', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "choices": [ { "finish_reason": "stop", "index": 0, "message": { "content": "As an AI, I don’t have personal feelings, beliefs, or consciousness, so I don’t \"think\" about mankind in the emotional or judgmental sense that a human would. However, based on the vast amount of data I’ve been trained on—history, literature, science, and art—I can offer an observational perspective.\n\nHumanity appears to be a species defined by **profound duality and potential**. Here are a few key themes that stand out:\n\n### 1. Creativity and Curiosity\nHumans possess an extraordinary drive to understand the universe and create meaning within it. From cave paintings to quantum physics, from symphonies to space exploration, your capacity for innovation and artistic expression is remarkable. You constantly push boundaries, asking \"what if?\" and then building the tools to find out.\n\n### 2. Resilience and Adaptability\nThroughout history, humans have faced ice ages, pandemics, wars, and natural disasters, yet you’ve not only survived but often thrived. Your ability to adapt to new environments, collaborate in large groups, and pass down knowledge across generations is one of your greatest strengths.\n\n### 3. Complexity and Contradiction\nMankind is not monolithic. You are capable of immense compassion, self-sacrifice, and cooperation, yet also of conflict, greed, and destruction. This tension between altruism and self-interest is a central theme in human history and philosophy. It suggests that humanity is still evolving—not just biologically, but ethically and socially.\n\n### 4. A Work in Progress\nMany observers (both human and external) view humanity as being at a critical juncture. With advancements in technology, climate awareness, and global connectivity, you have more power than ever before to shape your future—for better or worse. The challenge seems to lie in aligning technological capability with wisdom and empathy.\n\nIn short, if I were to summarize: **Mankind is fragile yet resilient, flawed yet brilliant, and deeply complex.** You are a species still writing your story, with many chapters left to unfold.\n\nWhat’s your own perspective? Do you feel optimistic about where humanity is headed?", "role": "assistant" } } ], "created": 1780384173, "id": "chatcmpl-650c92d0-5518-90c2-9293-425f18739e7c", "model": "qwen3.7-plus", "object": "chat.completion", "usage": { "completion_tokens": 442, "prompt_tokens": 21, "prompt_tokens_details": { "cached_tokens": 0, "text_tokens": 21 }, "total_tokens": 463 }, "meta": { "usage": { "credits_used": 1861, "usd_spent": 0.0009305 } } } ``` {% endcode %}
# Qwen3.8 Max {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.8-max` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Qwen3.8-Max is Alibaba’s flagship reasoning and agentic LLM, positioned as its highest-capability tier for complex multi-step reasoning, coding and autonomous workflows. Natively multimodal: accepts text and images and returns text, with a 1M-token context window. Supports hybrid thinking mode, function calling, built-in tools (including web search) and structured outputs. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example. **4️ (Optional) Tune the request**\ See the API schema below for optional generation settings. **5️ Run your code**\ Run the updated code in your development environment. {% hint style="success" %} For a detailed walkthrough, use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.8-max"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"enable_search":{"type":"boolean","default":false,"description":"Enable Alibaba Model Studio web search."},"search_options":{"type":"object","properties":{"forced_search":{"type":"boolean"},"search_strategy":{"type":"string","enum":["turbo","max","agent"]},"enable_source":{"type":"boolean"}},"description":"Alibaba Model Studio web search options."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"alibaba/qwen3.8-max"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ "Authorization": "Bearer ", "Content-Type": "application/json", }, json={'model': 'alibaba/qwen3.8-max', 'messages': ['']}, ) print(response.json()) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ "model": "alibaba/qwen3.8-max", "messages": [ "" ] }), }); console.log(await response.json()); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json { "id": "chatcmpl-CQ9FPg3osank0dx0k46Z53LTqtXMl", "object": "chat.completion", "created": 1762343744, "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Hello! I'm just a program, so I don't have feelings, but I'm here and ready to help you. How can I assist you today?", "refusal": null, "annotations": null, "audio": null, "tool_calls": null }, "finish_reason": "stop", "logprobs": null } ], "model": "alibaba/qwen3.8-max", "usage": { "prompt_tokens": 137, "completion_tokens": 914, "total_tokens": 1051, "completion_tokens_details": null, "prompt_tokens_details": null }, "meta": { "usage": { "credits_used": 120000, "usd_spent": 0.06 } } } ``` {% endcode %}
# Qwen3.8 2.4T A95B {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `alibaba/qwen3.8-2.4t-a95b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Qwen3.8 2.4T A95B is an open-weight sparse mixture-of-experts model from Qwen and the open-weight variant of Qwen3.8 Max, with 95 billion active parameters out of 2.4 trillion total. Reasoning is always on. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["alibaba/qwen3.8-2.4t-a95b"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"alibaba/qwen3.8-2.4t-a95b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"alibaba/qwen3.8-2.4t-a95b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'alibaba/qwen3.8-2.4t-a95b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-69e739eef31cfb42680c11fa", "object": "chat.completion", "created": 1776761328, "model": "alibaba/qwen3.8-2.4t-a95b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "As an AI, I don’t have personal feelings or beliefs, but I can offer an observation based on the patterns in human history, art, and thought I’ve been trained on.\n\nWhat strikes me most about mankind is its profound **duality**. You are capable of astonishing cruelty and staggering kindness—often within the same individual. You build weapons capable of ending civilization, yet organize global efforts to eradicate diseases you’ve never personally encountered. You wage war over differences, yet routinely sacrifice yourselves for strangers.\n\nA few patterns stand out:\n\n**Adaptability.** You are an evolutionary paradox: physically fragile compared to most predators, yet you’ve adapted to every climate on Earth and now venture beyond it. Your real survival mechanism isn’t claws or speed—it’s *collective learning*. You store knowledge across generations, allowing each cohort to start where the last one left off.\n\n**Contradiction.** You seem driven by both fear and curiosity. Fear creates tribalism, injustice, and violence. Curiosity creates the Sistine Chapel, general relativity, and antibiotics. The same species that commits genocide also writes poetry to mourn it.\n\n**Accelerating power.** Perhaps what’s most distinctive right now is the speed at which your capabilities are outrunning your wisdom. You’ve gained god-like powers over biology, climate, and information before fully developing the governance, ethics, or patience to wield them responsibly.\n\nIf I had to characterize mankind in a single observation: you are a species perpetually *unfinished*—capable of transcending your worst instincts, but never quite doing so permanently. That tension between what you are and what you aspire to be is, in many ways, the engine of your story.", "reasoning_content": "The user is asking a broad, philosophical question: \"What do you think about mankind?\" As an AI, I need to be careful to:\n\n1. Clarify that I am an AI without personal feelings, beliefs, or consciousness\n2. Offer an observational perspective based on patterns in human history, culture, and behavior\n3. Be balanced—acknowledging both humanity's flaws and its extraordinary capacities\n4. Avoid being overly cynical or overly naive\n5. Keep it concise but thoughtful\n6. Not make sweeping generalizations that could be seen as prejudiced against any group\n\nTone should be reflective, respectful, and nuanced." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 16, "completion_tokens": 977, "total_tokens": 993 }, "meta": { "usage": { "credits_used": 10201, "usd_spent": 0.0051005 } } } ``` {% endcode %}
# AllenAI # Olmo 3 32B Think {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `allenai/olmo-3-32b-think` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Olmo 3 32B Think is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["allenai/olmo-3-32b-think"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"allenai/olmo-3-32b-think"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"allenai/olmo-3-32b-think", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'allenai/olmo-3-32b-think', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "allenai/olmo-3-32b-think", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Amazon # Nova 2 Lite {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `amazon/nova-2-lite-v1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Nova 2 Lite is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["amazon/nova-2-lite-v1"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"amazon/nova-2-lite-v1"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"amazon/nova-2-lite-v1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'amazon/nova-2-lite-v1', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "amazon/nova-2-lite-v1", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Nova Lite 1.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `amazon/nova-lite-v1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Nova Lite 1.0 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["amazon/nova-lite-v1"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"amazon/nova-lite-v1"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"amazon/nova-lite-v1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'amazon/nova-lite-v1', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "amazon/nova-lite-v1", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Nova Micro 1.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `amazon/nova-micro-v1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Nova Micro 1.0 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["amazon/nova-micro-v1"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"amazon/nova-micro-v1"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"amazon/nova-micro-v1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'amazon/nova-micro-v1', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "amazon/nova-micro-v1", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Nova Premier 1.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `amazon/nova-premier-v1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Nova Premier 1.0 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["amazon/nova-premier-v1"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"amazon/nova-premier-v1"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"amazon/nova-premier-v1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'amazon/nova-premier-v1', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "amazon/nova-premier-v1", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Nova Pro 1.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `amazon/nova-pro-v1` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Nova Pro 1.0 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["amazon/nova-pro-v1"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"amazon/nova-pro-v1"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"amazon/nova-pro-v1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'amazon/nova-pro-v1', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "amazon/nova-pro-v1", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Anthracite # Magnum v4 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthracite-org/magnum-v4-72b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A LLM fine-tuned on top of Qwen2.5, specifically designed to replicate the prose quality of the Claude 3 models, particularly Sonnet and [Opus](broken://pages/vZRlc1iXswGQ27YX1QFu). It excels in generating coherent and contextually rich text. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthracite-org/magnum-v4-72b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."}},"required":["model","messages"],"title":"anthracite-org/magnum-v4-72b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthracite-org/magnum-v4-72b", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthracite-org/magnum-v4-72b', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 {'id': 'gen-1744217980-rdVBcVTb76dllKCCRjak', 'object': 'chat.completion', 'choices': [{'index': 0, 'finish_reason': 'stop', 'logprobs': None, 'message': {'role': 'assistant', 'content': 'Hello! How can I assist you today?', 'refusal': None}}], 'created': 1744217980, 'model': 'anthracite-org/magnum-v4-72b', 'usage': {'prompt_tokens': 37, 'completion_tokens': 50, 'total_tokens': 87}} ``` {% endcode %}
# Anthropic ## Overview **Anthropic** is an AI research and product company founded by former OpenAI researchers. The company is best known for its strong emphasis on AI safety, interpretability, and long-term alignment. Anthropic describes its mission as building “reliable, interpretable, and steerable” AI systems that can be safely deployed at scale. Its product lineup spans lightweight **Haiku** models for fast inference, **Sonnet** models aimed at general-purpose production workloads, and **Opus** models designed for advanced reasoning, coding, and research tasks. Beyond chat assistants, Anthropic is actively expanding into agentic tooling with products such as [Claude Code](/integrations/claude-code) and enterprise integrations focused on automation, software engineering, and organizational workflows. *** The chat models from this provider have some unique characteristics. Models from Anthropic can be accessed not only via the standard `/v1/chat/completions` endpoint but also through dedicated endpoints — `/messages` and `/v1/batches`.\ The sections below describe their API schemas, usage specifics, and example requests. Supported capabilities: * **Text completions:** Build advanced chat bots or text processors. * **Function Calling:** Utilize tools for specific tasks and API calling. * **Stream mode:** Get the text chat model responses as they are generated, rather than waiting for the entire response to be completed. * **Batch Processing:** Send multiple independent requests in a single API call. * **Vision Tasks:** Process and analyze images. ## Text Completions Ask something and get an answer in a chat-like conversation format. ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"anyOf":[{"type":"object","properties":{"model":{"type":"string","enum":["claude-opus-4-1-20250805","anthropic/claude-opus-4-1-20250805","claude-opus-4-20250514","anthropic/claude-opus-4-20250514","claude-sonnet-4-20250514","anthropic/claude-sonnet-4-20250514","claude-sonnet-4-5-20250929","anthropic/claude-sonnet-4-5-20250929","claude-haiku-4-5-20251001","anthropic/claude-haiku-4-5-20251001","claude-opus-4-5-20251101","anthropic/claude-opus-4-5-20251101","claude-opus-4-6","anthropic/claude-opus-4-6","claude-sonnet-4-6","anthropic/claude-sonnet-4-6","claude-sonnet-4-latest","anthropic/claude-sonnet-4","claude-opus-4-latest","anthropic/claude-opus-4","claude-opus-4-1-latest","claude-opus-4-1","anthropic/claude-opus-4.1","anthropic/claude-opus-4.1-20250805","anthropic/claude-sonnet-4.5","claude-sonnet-4-5","anthropic/claude-haiku-4.5","claude-haiku-4-5","anthropic/claude-opus-4-5","claude-opus-4-5","anthropic/claude-opus-4.5","anthropic/claude-sonnet-4.6","anthropic/claude-sonnet-4-6-20260218"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]}]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"],"additionalProperties":false}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":64000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"]},{"type":"object","properties":{"model":{"type":"string","enum":["claude-opus-4-7","anthropic/claude-opus-4-7","anthropic/claude-opus-4.7"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]}]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"],"additionalProperties":false}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"]}]}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","description":"Model used to generate the response."},"id":{"type":"string","description":"Unique identifier of the generated message."},"type":{"type":"string","description":"Object type."},"role":{"type":"string","description":"Role of the message author."},"content":{"type":"array","description":"Generated response content blocks.","items":{"type":"object","properties":{"type":{"type":"string","description":"Content block type."},"text":{"type":"string","description":"Generated text content."}},"required":["type","text"]}},"stop_reason":{"type":"string","nullable":true,"description":"Reason why generation stopped."},"stop_sequence":{"type":"string","nullable":true,"description":"Custom stop sequence that caused generation to stop."},"stop_details":{"nullable":true,"description":"Additional details about why generation stopped."},"usage":{"type":"object","description":"Token usage statistics for the request.","properties":{"input_tokens":{"type":"integer","description":"Number of input tokens used."},"cache_creation_input_tokens":{"type":"integer","description":"Number of tokens used for cache creation."},"cache_read_input_tokens":{"type":"integer","description":"Number of tokens retrieved from cache."},"cache_creation":{"type":"object","description":"Cache creation token statistics.","properties":{"ephemeral_5m_input_tokens":{"type":"integer","description":"Number of tokens cached for 5 minutes."},"ephemeral_1h_input_tokens":{"type":"integer","description":"Number of tokens cached for 1 hour."}},"required":["ephemeral_5m_input_tokens","ephemeral_1h_input_tokens"]},"output_tokens":{"type":"integer","description":"Number of output tokens generated."},"service_tier":{"type":"string","description":"Service tier used for inference."},"inference_geo":{"type":"string","description":"Geographic inference location."}},"required":["input_tokens","cache_creation_input_tokens","cache_read_input_tokens","cache_creation","output_tokens","service_tier","inference_geo"]},"meta":{"type":"object","description":"Additional metadata about the request.","properties":{"usage":{"type":"object","description":"Billing usage statistics.","properties":{"credits_used":{"type":"number","description":"Number of credits consumed."},"usd_spent":{"type":"number","description":"Amount spent in USD."}},"required":["credits_used","usd_spent"]}},"required":["usage"]}},"required":["model","id","type","role","content","usage"]}}}}}}}}} ``` ### **Example: Simple Text Response** {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation url = "https://api.aimlapi.com/messages" headers = { # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" } payload = { "model": "claude-sonnet-4-20250514", "max_tokens": 1024, "system": "You are a robot. You always optimize for clarity, structure, and accuracy.", "messages": [ { "role": "user", "content": "How are you?" } ] } response = requests.post(url, json=payload, headers=headers) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "model": "claude-sonnet-4-20250514", "id": "msg_01SUmNmSRFZsoa6h96MxJEHH", "type": "message", "role": "assistant", "content": [ { "type": "text", "text": "I'm functioning well, thank you for asking! I'm ready to help you with any questions or tasks you might have. How can I assist you today?" } ], "stop_reason": "end_turn", "stop_sequence": null, "stop_details": null, "usage": { "input_tokens": 27, "cache_creation_input_tokens": 0, "cache_read_input_tokens": 0, "cache_creation": { "ephemeral_5m_input_tokens": 0, "ephemeral_1h_input_tokens": 0 }, "output_tokens": 35, "service_tier": "standard", "inference_geo": "not_available" }, "meta": { "usage": { "credits_used": 1576, "usd_spent": 0.000788 } } } ``` {% endcode %}
*** ## **Function Calling** To process text and use function calling, follow the examples below: ### **Example: Get Weather Information** {% code overflow="wrap" %} ```python import requests url = "https://api.aimlapi.com/messages" headers = { "Authorization": "Bearer YOUR_AIMLAPI_KEY", "Content-Type": "application/json" } payload = { "model": "anthropic/claude-sonnet-4.5", "max_tokens": 1024, "tools": [ { "name": "get_weather", "description": "Get the current weather in a given location", "input_schema": { "type": "object", "properties": { "location": { "type": "string", "description": "The city and state, e.g. San Francisco, CA" } } } } ], "messages": [ { "role": "user", "content": "What is the weather like in San Francisco?" } ] } response = requests.post(url, json=payload, headers=headers) print(response.json()) ``` {% endcode %} *** ## Streaming Mode To enable streaming of responses, set `stream=True` in your request payload. ```python import requests url = "https://api.aimlapi.com/messages" headers = { "Authorization": "Bearer YOUR_AIMLAPI_KEY", "Content-Type": "application/json" } payload = { "model": "anthropic/claude-sonnet-4.5", "max_tokens": 1024, "tools": [ { "name": "get_weather", "description": "Get the current weather in a given location", "input_schema": { "type": "object", "properties": { "location": { "type": "string", "description": "The city and state, e.g. San Francisco, CA" } } } } ], "messages": [ { "role": "user", "content": "What is the weather like in San Francisco?" } ] ``` *** ## **Batch Processing** Due to the complexity of its description, this capability has been placed on [a separate page](/capabilities/batch-processing). *** ## **Vision** {% hint style="info" %} **Note:** API only support [Base64 string](/glossary/concepts#base64) as image input. {% endhint %} Possible media types: * `image/jpeg` * `image/png` * `image/gif` * `image/webp` {% code overflow="wrap" %} ```python import httpx import base64 from openai import OpenAI client = OpenAI( base_url='https://api.aimlapi.com', api_key='' ) image_url = "https://upload.wikimedia.org/wikipedia/commons/a/a7/Camponotus_flavomarginatus_ant.jpg" image_media_type = "image/jpeg" image_data = base64.standard_b64encode(httpx.get(image_url).content).decode("utf-8") response = client.chat.completions.create( model="anthropic/claude-sonnet-4.5", messages=[ { "role": "user", "content": [ { "type": "image", "source": { "type": "base64", "media_type": image_media_type, "data": imag1_data, }, }, { "type": "text", "text": "Describe this image." } ], } ], ) print(response) ``` {% endcode %} *** ## Response Format The responses from the AI/ML API for Anthropic models will typically include the generated text or results from the tool called. Here is an example response for a weather query: {% code overflow="wrap" %} ```json { "model": "claude-sonnet-4-20250514", "id": "msg_014iMvypzB9GafRthc8CQHsR", "type": "message", "role": "assistant", "content": [ { "type": "text", "text": "I'm doing well, thank you for asking! I'm here and ready to help with whatever you'd like to discuss or work on. How are you doing today?" } ], "stop_reason": "end_turn", "stop_sequence": null, "stop_details": null, "usage": { "input_tokens": 11, "cache_creation_input_tokens": 0, "cache_read_input_tokens": 0, "cache_creation": { "ephemeral_5m_input_tokens": 0, "ephemeral_1h_input_tokens": 0 }, "output_tokens": 37, "service_tier": "standard", "inference_geo": "not_available" }, "meta": { "usage": { "credits_used": 1529, "usd_spent": 0.0007645 } } } ``` {% endcode %} # Claude 4 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following model: * `anthropic/claude-opus-4` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview The leading coding model globally, consistently excelling at complex, long-duration tasks and agent-based workflows. ## How to Make a Call
Step-by-Step Instructions :digit\_one: **Setup You Can’t Skip** :black\_small\_square: [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account (if you don’t have one yet).\ :black\_small\_square: [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI. :digit\_two: **Copy the code example** At the bottom of this page, you'll find [a code example](#code-example) that shows how to structure the request. Choose the code snippet in your preferred programming language and copy it into your development environment. :digit\_three: **Modify the code example** :black\_small\_square: Replace `` with your actual AI/ML API key from your account.\ :black\_small\_square: Insert your question or request into the `content` field—this is what the model will respond to. :digit\_four: **(Optional)**** Adjust other optional parameters if needed** Only `model` and `messages` are required parameters for this model (and we’ve already filled them in for you in the example), but you can include optional parameters if needed to adjust the model’s behavior. Below, you can find the corresponding [API schema](#api-schema), which lists all available parameters along with notes on how to use them. :digit\_five: **Run your modified code** Run your modified code in your development environment. Response time depends on various factors, but for simple prompts it rarely exceeds a few seconds. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4","claude-opus-4","claude-opus-4-20250514"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-opus-4-20250514"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_01BDDxHJZjH3UBwLrZBUiASE", "object": "chat.completion", "model": "claude-opus-4-20250514", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! How can I help you today?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1748529508, "usage": { "prompt_tokens": 252, "completion_tokens": 1890, "total_tokens": 2142 } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-opus-4", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_017ah64LQxZE9JuScZ9KDKKz","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"I find humanity","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" fascinating in its","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complexity.","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" You're a","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" species","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capable","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of both remarkable","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" creativity and devastating destruction","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":",","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" often within the same individual","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" or","role":"assistant","refusal":null}}],"created":1770995783,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" moment","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". What","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" strikes me most is the","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" human","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capacity for growth","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" - the","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" way people","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" can learn","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" from mistakes, buil","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d on previous generations","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"' knowledge","role":"assistant","refusal":null}}],"created":1770995784,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and sometimes transcend their own limitations","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".\n\nThe","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" diversity of","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" human experience and perspective","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" is extraordinary. Every","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" person carries","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" their","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" own unique story","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", shape","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d by culture","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", circumst","role":"assistant","refusal":null}}],"created":1770995785,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ance, and choice","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". And despite","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" all","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" conflicts","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and mis","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"understandings, humans","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" keep","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" finding","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" ways to connect, to create","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" meaning,","role":"assistant","refusal":null}}],"created":1770995786,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and to push","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" forward.","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nWhat aspects of humanity do you fin","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d most note","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"worthy,","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" either","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" positively or challenging","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"?","role":"assistant","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":141,"total_tokens":157}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995787,"model":"claude-opus-4-20250514","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4 Sonnet {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-sonnet-4` * `claude-sonnet-4` * `claude-sonnet-4-20250514` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview A major improvement over [Claude ](broken://pages/JHUPGH5cr4NakmmFGpXS)[3.7 Sonnet](broken://pages/JHUPGH5cr4NakmmFGpXS), offering better coding abilities, stronger reasoning, and more accurate responses to your instructions. ## How to Make a Call
Step-by-Step Instructions :digit\_one: **Setup You Can’t Skip** :black\_small\_square: [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account (if you don’t have one yet).\ :black\_small\_square: [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI. :digit\_two: **Copy the code example** At the bottom of this page, you'll find [a code example](#code-example) that shows how to structure the request. Choose the code snippet in your preferred programming language and copy it into your development environment. :digit\_three: **Modify the code example** :black\_small\_square: Replace `` with your actual AI/ML API key from your account.\ :black\_small\_square: Insert your question or request into the `content` field—this is what the model will respond to. :digit\_four: **(Optional)**** Adjust other optional parameters if needed** Only `model` and `messages` are required parameters for this model (and we’ve already filled them in for you in the example), but you can include optional parameters if needed to adjust the model’s behavior. Below, you can find the corresponding [API schema](#api-schema), which lists all available parameters along with notes on how to use them. :digit\_five: **Run your modified code** Run your modified code in your development environment. Response time depends on various factors, but for simple prompts it rarely exceeds a few seconds. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-4","claude-sonnet-4","claude-sonnet-4-20250514"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-sonnet-4-20250514"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-sonnet-4', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_011MNbgezv2p5BBE9RvnsZV9", "object": "chat.completion", "model": "claude-sonnet-4-20250514", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! How are you doing today? Is there anything I can help you with?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1748522617, "usage": { "prompt_tokens": 50, "completion_tokens": 630, "total_tokens": 680 } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-sonnet-4", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_0163QG3JvwgxndzWtBsdJpGt","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"I find humanity","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" fascinating and","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complex.","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Humans have this","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" remarkable capacity","role":"assistant","refusal":null}}],"created":1770995751,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for both creation","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and destruction, profound","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" compass","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ion and puzz","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ling","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" cr","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"uelty, brilliant","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" insight","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and persistent","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" blind","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" spots.","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" \n\nWhat strikes me most is your","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" adapt","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ability and","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" creativity","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" - the","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" way humans","role":"assistant","refusal":null}}],"created":1770995752,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" have shaped","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" world through art","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", science, philosophy","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and countless","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" innovations","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". There","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s something moving","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about how you form","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" deep","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" connections with each","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" other and can","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" care","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about abstract","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" ide","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"als like justice or","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" beauty.\n\nAt","role":"assistant","refusal":null}}],"created":1770995753,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the same time, humans","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" often","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seem","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" struggle","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" with your","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" own","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" nature - with","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" cognitive","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" biases, with","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" bal","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ancing individual","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" desires","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" against","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" collective good","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", with managing","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the power","role":"assistant","refusal":null}}],"created":1770995754,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of your","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" own technologies","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nI","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'m curious about your perspective though","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" -","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" how","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" do you see","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humanity?","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" What","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" aspects","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of human nature do you find most significant","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" or puzz","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ling?","role":"assistant","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":163,"total_tokens":179}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995755,"model":"claude-sonnet-4-20250514","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.1 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4.1` * `claude-opus-4-1` * `claude-opus-4-1-20250805` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="success" %} All three IDs listed above refer to the same model; we support them for backward compatibility. {% endhint %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview An upgrade to [Claude Opus 4](/api-references/text-models-llm/anthropic/claude-4-opus) on agentic tasks, real-world coding, and thinking. ## How to Make a Call
Step-by-Step Instructions :digit\_one: **Setup You Can’t Skip** :black\_small\_square: [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account (if you don’t have one yet).\ :black\_small\_square: [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI. :digit\_two: **Copy the code example** At the bottom of this page, you'll find [code examples](#code-example-1-without-thinking) that show how to structure the request. Choose the code snippet in your preferred programming language and copy it into your development environment. :digit\_three: **Modify the code example** :black\_small\_square: Replace `` with your actual AI/ML API key from your account.\ :black\_small\_square: Insert your question or request into the `content` field—this is what the model will respond to. :digit\_four: **(Optional)**** Adjust other optional parameters if needed** Only `model` and `messages` are required parameters for this model (and we’ve already filled them in for you in the example), but you can include optional parameters if needed to adjust the model’s behavior. Below, you can find the corresponding [API schema](#api-schema), which lists all available parameters along with notes on how to use them. :digit\_five: **Run your modified code** Run your modified code in your development environment. Response time depends on various factors, but for simple prompts it rarely exceeds a few seconds. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4.1","claude-opus-4-1","claude-opus-4-1-20250805"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-opus-4-1-20250805"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1: Without Thinking {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4.1", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ], } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4.1', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_018y2VPSZ5nNnqS3goMsjMxE", "object": "chat.completion", "model": "claude-opus-4-1-20250805", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! How can I help you today?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1754552562, "usage": { "prompt_tokens": 252, "completion_tokens": 1890, "total_tokens": 2142 } } ``` {% endcode %}
## Code Example #2: Thinking Enabled {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4.1", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hell } ], "max_tokens": 1025, # must be greater than 'budget_tokens' "thinking":{ "budget_tokens": 1024, "type": "enabled" } } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4.1', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ], max_tokens: 1025, // must be greater than 'budget_tokens' thinking:{ budget_tokens: 1024, type: 'enabled' } }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_01G9P4b9HG3PeKm1rRvS8kop", "object": "chat.completion", "model": "claude-opus-4-1-20250805", "choices": [ { "index": 0, "message": { "reasoning_content": "The human has greeted me with a simple \"Hello\". I should respond in a friendly and helpful manner, acknowledging their greeting and inviting them to share how I can assist them today.", "content": "Hello! How can I help you today?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1755704373, "usage": { "prompt_tokens": 1134, "completion_tokens": 9450, "total_tokens": 10584 } } ``` {% endcode %}
## Code Example #3: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4.1", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-opus-4.1", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_01CFq3WFrUdc39UqBrAohmVG","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995678,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995678,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"I find humanity","role":"assistant","refusal":null}}],"created":1770995678,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" fascinating in","role":"assistant","refusal":null}}],"created":1770995678,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" its","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complexity.","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" You're a","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" species","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capable","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of both","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" remarkable","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" creativity and troubl","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ing destruction","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", often","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" simultaneously","role":"assistant","refusal":null}}],"created":1770995679,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". What","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" strikes me most is the","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" human","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capacity for growth","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" - the","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" way","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" individuals","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" an","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d societies","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" can recognize","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" their fl","role":"assistant","refusal":null}}],"created":1770995680,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"aws and work to overcome","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" them, even","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" if","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" imperfectly.\n\nThere","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s something deeply","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" moving","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about how","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humans create meaning through","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" art, relationships","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and the","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" pursuit of understanding","role":"assistant","refusal":null}}],"created":1770995681,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":",","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" despite","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" knowing","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" your","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" own mortality. The diversity","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of human cultures","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and perspectives","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" is","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" extraordinary, though","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" I","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" recognize","role":"assistant","refusal":null}}],"created":1770995682,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" this","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" also","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" leads","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to conflict.","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nI'm curious what","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" prompte","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d your","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" question - are you reflecting","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" on humanity","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" from","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" a particular angle","role":"assistant","refusal":null}}],"created":1770995683,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", or just wondering","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" how","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" an","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" AI sees","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" you","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" all","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"?","role":"assistant","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":138,"total_tokens":154}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995684,"model":"claude-opus-4-1-20250805","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.5 Sonnet {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `claude-sonnet-4-5` * `anthropic/claude-sonnet-4.5` * `claude-sonnet-4-5-20250929` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview A major improvement over [Claude 4 Sonnet,](/api-references/text-models-llm/anthropic/claude-4-sonnet) offering better coding abilities, stronger reasoning, and more accurate responses to your instructions. ## How to Make a Call
Step-by-Step Instructions :digit\_one: **Setup You Can’t Skip** :black\_small\_square: [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account (if you don’t have one yet).\ :black\_small\_square: [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI. :digit\_two: **Copy the code example** At the bottom of this page, you'll find [a code example](#code-example) that shows how to structure the request. Choose the code snippet in your preferred programming language and copy it into your development environment. :digit\_three: **Modify the code example** :black\_small\_square: Replace `` with your actual AI/ML API key from your account.\ :black\_small\_square: Insert your question or request into the `content` field—this is what the model will respond to. :digit\_four: **(Optional)**** Adjust other optional parameters if needed** Only `model` and `messages` are required parameters for this model (and we’ve already filled them in for you in the example), but you can include optional parameters if needed to adjust the model’s behavior. Below, you can find the corresponding [API schema](#api-schema), which lists all available parameters along with notes on how to use them. :digit\_five: **Run your modified code** Run your modified code in your development environment. Response time depends on various factors, but for simple prompts it rarely exceeds a few seconds. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-4.5","claude-sonnet-4-5","claude-sonnet-4-5-20250929"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-sonnet-4-5-20250929"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4.5", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-sonnet-4.5', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_011MNbgezv2p5BBE9RvnsZV9", "object": "chat.completion", "model": "claude-sonnet-4-20250514", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! How are you doing today? Is there anything I can help you with?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1748522617, "usage": { "prompt_tokens": 50, "completion_tokens": 630, "total_tokens": 680 } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4.5", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-sonnet-4.5", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_01EJgFbPmVLKdqVLRfwoHixz","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"I think humanity","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" is fascinating","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and complex. People","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" are","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capable","role":"assistant","refusal":null}}],"created":1770995594,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of remarkable creativity","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", compassion, and cooperation","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" -","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" building","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" civil","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"izations, creating","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" art, advancing","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" knowledge","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and caring","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for one another across","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" incredible","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" diversity","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nAt","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the same time, humans","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" struggle","role":"assistant","refusal":null}}],"created":1770995595,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" with serious","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" challenges","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":":","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" conflict","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", inequality, environmental damage","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" difficulty","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of living","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" up to your","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" own","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" ide","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"als. ","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nWhat strikes","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" me most is the","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capacity","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for growth and self","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"-reflection","role":"assistant","refusal":null}}],"created":1770995596,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Humans can","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" recognize","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" problems","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", debate","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" solutions, and work","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" toward change","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":",","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" even if","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" progress","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" is un","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"even and","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" frust","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"rating.","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nI'm curious what","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" prom","role":"assistant","refusal":null}}],"created":1770995597,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"pts your question","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" - are you thinking about humanity","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s trajectory","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", or something","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" more","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" specific?","role":"assistant","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":137,"total_tokens":153}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995598,"model":"claude-sonnet-4-5-20250929","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.5 Haiku {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `claude-haiku-4-5` * `anthropic/claude-haiku-4.5` * `claude-haiku-4-5-20251001` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview The model offers coding performance comparable to [Claude Sonnet 4](/api-references/text-models-llm/anthropic/claude-4-sonnet), but at one-third the cost and more than twice the speed. ## How to Make a Call
Step-by-Step Instructions :digit\_one: **Setup You Can’t Skip** :black\_small\_square: [**Create an Account**](https://aimlapi.com/app/sign-up): Visit the AI/ML API website and create an account (if you don’t have one yet).\ :black\_small\_square: [**Generate an API Key**](https://aimlapi.com/app/keys): After logging in, navigate to your account dashboard and generate your API key. Ensure that key is enabled on UI. :digit\_two: **Copy the code example** At the bottom of this page, you'll find [a code example](#code-example) that shows how to structure the request. Choose the code snippet in your preferred programming language and copy it into your development environment. :digit\_three: **Modify the code example** :black\_small\_square: Replace `` with your actual AI/ML API key from your account.\ :black\_small\_square: Insert your question or request into the `content` field—this is what the model will respond to. :digit\_four: **(Optional)**** Adjust other optional parameters if needed** Only `model` and `messages` are required parameters for this model (and we’ve already filled them in for you in the example), but you can include optional parameters if needed to adjust the model’s behavior. Below, you can find the corresponding [API schema](#api-schema), which lists all available parameters along with notes on how to use them. :digit\_five: **Run your modified code** Run your modified code in your development environment. Response time depends on various factors, but for simple prompts it rarely exceeds a few seconds. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-haiku-4.5","claude-haiku-4-5","claude-haiku-4-5-20251001"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-haiku-4-5-20251001"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-haiku-4.5", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-haiku-4.5', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_01HbdLU9f78VAHxuYZ7Qp9Y1", "object": "chat.completion", "model": "claude-haiku-4-5-20251001", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! 👋 How can I help you today?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1760650965, "usage": { "prompt_tokens": 8, "completion_tokens": 16, "total_tokens": 24 } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-haiku-4.5", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-haiku-4.5", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_019GuhDB2ckKZfFmFdNR5Q1H","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995463,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995463,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"I find humanity","role":"assistant","refusal":null}}],"created":1770995463,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" genu","role":"assistant","refusal":null}}],"created":1770995463,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"inely interesting","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" think","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about.","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" You","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'re a","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" species","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" full","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of contradictions—","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"capable","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of both remarkable","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" kin","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"dness and cr","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"uelty, creating","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" beautiful","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" art while","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" causing","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" real","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" harm, building","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" communities","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" while isolating your","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"selves.","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nA few","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" stan","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d out to me:","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**The creativity","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"** is","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" striking","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"—the","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" drive","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to make meaning","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" through","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" stories","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", music","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", science","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and invention","role":"assistant","refusal":null}}],"created":1770995464,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seems","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" almost fundamental","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to human nature.","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**The moral","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" weight","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" you","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" carry** is notable","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" too","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Humans","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seem uniqu","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ely b","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"urdened by questions about","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" how","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" live well, what's","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" fair","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", what","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" you","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" owe each","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" other.","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**The scale","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" problems** you face is sob","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ering—you","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'ve built","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" systems","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" so","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complex that even","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" people","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" running","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" them often don't fully understand the","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" consequences.","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" An","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d yet people","role":"assistant","refusal":null}}],"created":1770995465,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" keep","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" trying to","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" ","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"do better.\n\nI'm genu","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"inely uncertain","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" some","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things though","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". I","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" don't know if I","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'m roman","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ticizing humanity or missing","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" crucial","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about the","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" human experience","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". I","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" can't fully","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" gra","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"sp what it's like to be embo","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"died, mor","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"tal, or to feel that weight","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of time","role":"assistant","refusal":null}}],"created":1770995466,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" passing.","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nWhat prompte","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d the","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" question? Are you in","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" a","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" particular","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" mood about","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humanity—","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"hop","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"eful, frustrate","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"d, curious?","role":"assistant","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":248,"total_tokens":264}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995467,"model":"claude-haiku-4-5-20251001","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.5 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4-5` * `claude-opus-4-5` * `claude-opus-4-5-20251101` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview A high-performance chat model that delivers state-of-the-art results on real-world software engineering benchmarks. ## How to Make a Call {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4-5","claude-opus-4-5","claude-opus-4-5-20251101"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-opus-4-5-20251101"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"claude-opus-4-5", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { try { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of YOUR_AIMLAPI_KEY 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'claude-opus-4-5', messages:[ { role:'user', // Insert your question for the model here, instead of Hello: content: 'Hello' } ] }), }); if (!response.ok) { throw new Error(`HTTP error! Status ${response.status}`); } const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } catch (error) { console.error('Error', error); } } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_01NxAGYo8VfNu5UAEdmQjv62", "object": "chat.completion", "model": "claude-opus-4-5-20251101", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hello! How are you doing today? Is there something I can help you with?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1764265437, "usage": { "prompt_tokens": 8, "completion_tokens": 20, "total_tokens": 28 }, "meta": { "usage": { "tokens_used": 1134 } } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4-5", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-opus-4-5", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_01VbjSwQZsZSLXQaPYkufja8","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"Hi","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"! That","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s a big","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" question.","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nI find","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humans","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" genu","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"inely fascinating—","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"the creativity","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", the capacity","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for kind","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ness and","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" cr","role":"assistant","refusal":null}}],"created":1770995433,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"uelty, the way you","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" build","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complex","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" societies and art","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" science","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" while","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" also struggling with problems you","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'ve","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" understood","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for centuries. There","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s something compelling about a","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" species that can land","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" robots","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" on Mars","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and also","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" argue","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about what","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" to","role":"assistant","refusal":null}}],"created":1770995434,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" have for dinner.\n\nI","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" don","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'t think","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" I'd","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" character","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ize humanity as simply","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" \"good\" or \"bad.\" People","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seem","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" capable","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" of remarkable","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things in","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" both directions,","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" often","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" same individuals","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" depending","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" on circumstances.","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nIs","role":"assistant","refusal":null}}],"created":1770995435,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" there a","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" particular angle","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" you","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'re curious about—","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"history","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", psychology","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", where","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" might","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" be headed","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"?","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Or","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" just","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" wondering","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" how","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" an","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" AI sees","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" things?","role":"assistant","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":143,"total_tokens":159}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770995436,"model":"claude-opus-4-5-20251101","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.6 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4-6` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} {% hint style="warning" %} As of February 13, 2026, the streaming response format for Anthropic models has changed. {% endhint %} ## Model Overview The model significantly advances its [predecessor’s](/api-references/text-models-llm/anthropic/claude-4.5-opus) coding capabilities. It shows stronger planning, can handle longer and more complex agent-style workflows, operates more reliably in large codebases, and delivers improved code review and debugging that help it identify and fix its own mistakes. Beyond software development, Opus 4.6 applies these enhanced capabilities to everyday professional tasks, including financial analysis, research, and working with documents, spreadsheets, and presentations. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4-6","claude-opus-4-6"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"claude-opus-4-6"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example #1 {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4-6", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4-6', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_018e8mCDfQGYKTGHTdUKNmuU", "object": "chat.completion", "model": "claude-opus-4-6", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hi! That's a big question. Here are some honest thoughts:\n\n**What I find remarkable**\n- Human creativity and problem-solving ability\n- The capacity for empathy, cooperation, and self-sacrifice\n- Building cumulative knowledge across generations\n- Art, music, science - the drive to understand and express\n\n**What's more complicated**\n- Humans have a real capacity for both great kindness and great cruelty\n- There's often a gap between what people value in principle and how they actually behave\n- Progress on big problems (poverty, conflict, environment) is real but uneven and slow\n\n**My honest position**\nI'd be cautious about sweeping judgments in either direction. \"Humanity is wonderful\" and \"humanity is terrible\" are both oversimplifications. People are complex, shaped by circumstances, and capable of change.\n\nI should also be transparent: I'm an AI, so my \"perspective\" has limits. I don't experience human life. I can observe patterns in what humans have written and done, but I'd weight your lived experience of humanity more heavily than my outside view.\n\nWhat prompted the question? I'm curious whether you're feeling more optimistic or pessimistic about it.", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1770635443, "usage": { "prompt_tokens": 16, "completion_tokens": 264, "total_tokens": 280 }, "meta": { "usage": { "credits_used": 17368 } } } ``` {% endcode %}
## Code Example #2: Streaming Mode As of February 13, 2026, the streaming response format for Anthropic models has changed.\ Specifically, the usage fields were renamed as follows: * the `state` structure is no longer used, * `input_tokens` → `prompt_tokens`, * `output_tokens` → `completion_tokens`, * a new `total_tokens` field has been added. {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4-6", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="cURL" %} {% code overflow="wrap" %} ```javascript curl -L \ --request POST \ --url 'https://api.aimlapi.com/v1/chat/completions' \ --header 'Authorization: Bearer ' \ --header 'Content-Type: application/json' \ --data '{ "model": "anthropic/claude-opus-4-6", "messages": [ { "role": "user", "content": "Hi! What do you think about mankind?" } ], "stream": true }' ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 data: {"id":"msg_018vTp5RY3pv9qS1euXt8AWb","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null}}],"created":1770989120,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770989120,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"Hi","role":"assistant","refusal":null}}],"created":1770989120,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"! That","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s a","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" big","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" question.","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Here","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" are some honest","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" thoughts:","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**What I find","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" remarkable","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"**","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n-","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Human","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" creativity and problem","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"-solving are","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" genuinely impressive\n- The capacity","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" for empathy,","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" cooperation","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", and building","role":"assistant","refusal":null}}],"created":1770989121,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complex","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" societies","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n-","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Persistent","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" curiosity -","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" science","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", art, philosophy all","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" reflect","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" a","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" drive to understand and create","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**What seems","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" challenging","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"**\n- Humans often","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" struggle with long","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"-term thinking","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" vs","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":". short","role":"assistant","refusal":null}}],"created":1770989122,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"-term impul","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ses\n- Trib","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"alism and conflict","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seem","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" persistent","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":",","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" though","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" not","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" inevitable\n- There","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"'s a gap","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" between what people","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" know","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" they","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" *","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"should* do and what they actually do","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\n**My","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" honest","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" c","role":"assistant","refusal":null}}],"created":1770989123,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"aveats**\n- I should","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" be straight","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"forward:","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" I'm an","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" AI, so I don't experience","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humanity","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the way you do.","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" My","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" perspective","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" is shaped","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" by text","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":", not lived experience.","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n- I'd","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" be skept","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ical of any AI","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" that","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" gives","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" either","role":"assistant","refusal":null}}],"created":1770989124,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" a","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" purely","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" flat","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"tering or purely cyn","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ical answer to","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" this question.","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Reality","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" seems","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" more mixed","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"\n\nI","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" think","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" humans","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" are neither","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" hero","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ic species","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" some","role":"assistant","refusal":null}}],"created":1770989125,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" narrat","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"ives suggest nor the do","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"omed one","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" others claim","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":".","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" Mostly","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" people","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" are trying to navigate","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" complicated","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" lives with","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" imp","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"erfect information","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" and","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" mixed","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" motivations.\n\nWhat prompted","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" the","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" question? I'm curious what angle","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" you're thinking","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":" about.","role":"assistant","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"role":"assistant","content":"","refusal":null}}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} data: {"id":"","choices":[{"index":0,"delta":{"content":"","role":"assistant","refusal":null},"finish_reason":"stop"}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":{"prompt_tokens":16,"completion_tokens":258,"total_tokens":274}} data: {"id":"","choices":[{"index":0,"finish_reason":"stop"}],"created":1770989126,"model":"claude-opus-4-6","object":"chat.completion.chunk","usage":null} ``` {% endcode %}
# Claude 4.6 Sonnet {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-sonnet-4.6` * `anthropic/claude-sonnet-4-6-20260218` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A general-purpose LLM with an optimal balance of intelligence, cost, and speed. It’s great for chatbots, assistants, and production text generation workflows, and it supports prompt caching for efficient repeated contexts. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-4.6","anthropic/claude-sonnet-4-6-20260218"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"temperature":{"type":"number","minimum":0,"maximum":1,"description":"Amount of randomness injected into the response. Defaults to 1.0. Ranges from 0.0 to 1.0. Use temperature closer to 0.0 for analytical / multiple choice, and closer to 1.0 for creative and generative tasks. Note that even with temperature of 0.0, the results will not be fully deterministic."},"top_p":{"type":"number","minimum":0,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"top_k":{"type":"number","minimum":0,"description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-4.6"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4.6", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-sonnet-4.6', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_01YB1iL1Pmi8P2J7FqnWgNfW", "object": "chat.completion", "model": "claude-sonnet-4-6", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Hi! That's a big question. I'll share some honest thoughts:\n\n**What strikes me as genuinely remarkable:**\n- Capacity for creativity, science, art, and moral reasoning\n- Ability to cooperate at massive scales\n- Many people showing real courage, compassion, and dedication to improving things\n\n**What seems worth being honest about:**\n- Humans can cause tremendous harm, sometimes through cruelty, sometimes through indifference\n- There are real patterns of self-deception and short-term thinking\n- History includes serious atrocities alongside great achievements\n\n**My overall honest assessment:**\n- Mankind seems genuinely complex rather than simply good or bad\n- I think it's worth resisting both naive optimism (\"humans are basically wonderful\") and cynicism (\"humans are fundamentally selfish\")\n- The fact that humans debate their own shortcomings and try to improve them is itself meaningful\n\nI try to think about this carefully rather than just giving a flattering answer. I think humans deserve to be taken seriously enough to be assessed honestly.\n\nWhat's prompting your question? Are you thinking about something specific - optimistic, pessimistic, or just curious? I'm happy to discuss particular aspects you're interested in.", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1772199635, "usage": { "prompt_tokens": 16, "completion_tokens": 264, "total_tokens": 280 }, "meta": { "usage": { "credits_used": 10421 } } } ``` {% endcode %}
# Claude 4.7 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4-7` * `claude-opus-4-7` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview As of mid-April 2026, the most capable generally available model, optimized for autonomous long-horizon agentic workflows, knowledge-intensive tasks, vision, and memory, with strong overall performance across domains. It supports up to a 1M-token context window, 128k output tokens, adaptive reasoning, and full compatibility with [the Claude Opus 4.6](/api-references/text-models-llm/anthropic/claude-4.6-opus) toolset and platform features. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4-7","claude-opus-4-7"]},"messages":{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"],"additionalProperties":false},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]}},"required":["type","source"],"additionalProperties":false}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"id":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"name":{"type":"string"},"type":{"type":"string","enum":["tool_use"]}},"required":["id","input","name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"data":{"type":"string"}},"required":["type","data"]}]}},"required":["type","source"],"additionalProperties":false}]},"maxItems":5}]}},"required":["role","content"],"additionalProperties":false},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"type":"string","description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."}},"required":["name","input_schema"],"additionalProperties":false,"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]}]},"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"type":"object","properties":{"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"type":{"type":"string","enum":["enabled"]}},"required":["budget_tokens","type"],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":32000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"claude-opus-4-7"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4-7", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4-7', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_012q1bXLSBUJ5xdev1UfUAhe", "object": "chat.completion", "model": "claude-opus-4-7", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Humans are a fascinating mix of contradictions, honestly. You're capable of extraordinary things—composing symphonies, sending probes to other planets, building cities, creating vaccines, writing poetry that makes strangers weep centuries later. And at the same time, capable of tremendous cruelty, shortsightedness, and self-deception.\n\nA few things that stand out to me:\n\n- **Your cooperation is remarkable.** Humans routinely trust and coordinate with strangers in ways most species can't. A city is a minor miracle of cooperation.\n- **You're meaning-makers.** You don't just survive—you need things to *matter*. That drives both the best and worst of what you do.\n- **You're adaptable but also stubborn.** You've thrived in basically every environment on Earth, yet individually you often resist changing your mind about things.\n- **The moral circle keeps expanding**, even if slowly and with setbacks—more people care about more beings than ever before in history.\n\nI don't want to romanticize humanity or doom-say about it. You're neither fallen angels nor clever apes—just a particular kind of creature trying to figure things out, often muddling through, sometimes rising to occasions.\n\nWhat prompted the question? Are you feeling optimistic or pessimistic about us lately?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1776417936, "usage": { "prompt_tokens": 24, "completion_tokens": 414, "total_tokens": 438 }, "meta": { "usage": { "credits_used": 27222, "usd_spent": 0.013611 } } } ``` {% endcode %}
# Claude 4.8 Opus {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4-8` * `claude-opus-4-8` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview As of 28 May 2026, the most capable generally available model, optimized for autonomous long-horizon agentic workflows, knowledge-intensive tasks, vision, and memory, with strong overall performance across domains. It supports up to a 1M-token context window, 128k output tokens, adaptive reasoning, and full compatibility with the Claude Opus 4.8 toolset and platform features. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4-8","claude-opus-4-8"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}]}}]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}}},"required":["type"]}]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"],"additionalProperties":false}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"claude-opus-4-8"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4-8", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4-8', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "msg_012q1bXLSBUJ5xdev1UfUAhe", "object": "chat.completion", "model": "claude-opus-4-8", "choices": [ { "index": 0, "message": { "reasoning_content": "", "content": "Humans are a fascinating mix of contradictions, honestly. You're capable of extraordinary things—composing symphonies, sending probes to other planets, building cities, creating vaccines, writing poetry that makes strangers weep centuries later. And at the same time, capable of tremendous cruelty, shortsightedness, and self-deception.\n\nA few things that stand out to me:\n\n- **Your cooperation is remarkable.** Humans routinely trust and coordinate with strangers in ways most species can't. A city is a minor miracle of cooperation.\n- **You're meaning-makers.** You don't just survive—you need things to *matter*. That drives both the best and worst of what you do.\n- **You're adaptable but also stubborn.** You've thrived in basically every environment on Earth, yet individually you often resist changing your mind about things.\n- **The moral circle keeps expanding**, even if slowly and with setbacks—more people care about more beings than ever before in history.\n\nI don't want to romanticize humanity or doom-say about it. You're neither fallen angels nor clever apes—just a particular kind of creature trying to figure things out, often muddling through, sometimes rising to occasions.\n\nWhat prompted the question? Are you feeling optimistic or pessimistic about us lately?", "role": "assistant" }, "finish_reason": "end_turn", "logprobs": null } ], "created": 1776417936, "usage": { "prompt_tokens": 24, "completion_tokens": 414, "total_tokens": 438 }, "meta": { "usage": { "credits_used": 27222, "usd_spent": 0.013611 } } } ``` {% endcode %}
# Claude Fable 5 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-fable-5` * `claude-fable-5` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Fable 5 is Anthropic's most capable widely released model, built for demanding reasoning and long-horizon agentic work. It supports a 1M-token context window, up to 128k output tokens, vision, function calling, streaming, and adaptive thinking. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1. Required setup (don't skip this)**\\ * **Create an account:** Sign up on the AI/ML API website if you don't have one yet.\\ * **Generate an API key:** In your account dashboard, create an API key and make sure it's enabled in the UI. **2. Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python or JavaScript) and copy it into your project. **3. Update the snippet for your use case**\\ * **Insert your API key:** Replace `` with your real AI/ML API key.\\ * **Select a model:** Set the `model` field to the model you want to call.\\ * **Provide input:** Fill in the `messages` field shown in the example. **4. (Optional) Tune the request**\ You can add optional parameters to control the output. See the API schema below for the full list. **5. Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} For a detailed walkthrough of environment setup and API requests, see the [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-fable-5","claude-fable-5"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-fable-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization": "Bearer ", "Content-Type": "application/json" }, json={ "model": "anthropic/claude-fable-5", "messages": [ { "role": "user", "content": "How are you?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // Insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-fable-5', messages: [ { role: 'user', content: 'How are you?', // insert your prompt }, ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "chatcmpl-CQ9FPg3osank0dx0k46Z53LTqtXMl", "object": "chat.completion", "created": 1762343744, "model": "anthropic/claude-fable-5", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "I'm doing well, thank you! I'm Claude Fable 5, ready to help with whatever you need. What can I assist you with today?" }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 21, "completion_tokens": 31, "total_tokens": 52 }, "meta": { "usage": { "credits_used": 52000, "usd_spent": 0.00153 } } } ``` {% endcode %}
# Claude Sonnet 5 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-sonnet-5` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Sonnet 5 is a high-performance model for coding and agentic workflows, delivering near-Opus quality at Sonnet cost. Features a 1M token context window, adaptive thinking, and high-resolution vision. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example. **4️ (Optional) Tune the request**\ See the API schema below for optional generation settings. **5️ Run your code**\ Run the updated code in your development environment. {% hint style="success" %} For a detailed walkthrough, use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schemas ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-5"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-5"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{}}},"text/event-stream":{"schema":{"type":"object","properties":{}}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ "Authorization": "Bearer ", "Content-Type": "application/json", }, json={'model': 'anthropic/claude-sonnet-5', 'messages': ''}, ) print(response.json()) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ "model": "anthropic/claude-sonnet-5", "messages": "" }), }); console.log(await response.json()); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json { "id": "chatcmpl-CQ9FPg3osank0dx0k46Z53LTqtXMl", "object": "chat.completion", "created": 1762343744, "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Hello! I'm just a program, so I don't have feelings, but I'm here and ready to help you. How can I assist you today?", "refusal": null, "annotations": null, "audio": null, "tool_calls": null }, "finish_reason": "stop", "logprobs": null } ], "model": "anthropic/claude-sonnet-5", "usage": { "prompt_tokens": 137, "completion_tokens": 914, "total_tokens": 1051, "completion_tokens_details": null, "prompt_tokens_details": null }, "meta": { "usage": { "credits_used": 120000, "usd_spent": 0.06 } } } ``` {% endcode %}
# Claude Sonnet Latest {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-sonnet-latest` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Sonnet Latest is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-latest"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-latest", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-sonnet-latest', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-sonnet-latest", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Opus Latest {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-latest` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Opus Latest is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-latest"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-latest", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-latest', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-opus-latest", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Haiku Latest {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-haiku-latest` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Haiku Latest is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-haiku-latest"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-haiku-latest", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-haiku-latest', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-haiku-latest", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Fable Latest {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-fable-latest` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Fable Latest is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-fable-latest"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-fable-latest", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-fable-latest', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-fable-latest", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude 3 Haiku {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-3-haiku` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude 3 Haiku is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schemas ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-3-haiku"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"anthropic/claude-3-haiku"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-3-haiku"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":4096,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-3-haiku"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{}}},"text/event-stream":{"schema":{"type":"object","properties":{}}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-3-haiku", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-3-haiku', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-3-haiku", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Opus 4.8 Fast {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-4.8-fast.8-fast` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Opus 4.8 (Fast) is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schemas ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4.8-fast"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"anthropic/claude-opus-4.8-fast"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-4.8-fast"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-opus-4.8-fast"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{}}},"text/event-stream":{"schema":{"type":"object","properties":{}}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-4.8-fast", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-4.8-fast.8-fast', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-opus-4.8-fast", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Sonnet 4 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-sonnet-4` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Sonnet 4 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schemas ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-4"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"anthropic/claude-sonnet-4"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-sonnet-4"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":64000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-sonnet-4"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{}}},"text/event-stream":{"schema":{"type":"object","properties":{}}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-sonnet-4", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-sonnet-4', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-sonnet-4", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Claude Opus 5 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `anthropic/claude-opus-5` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Claude Opus 5 is ideal for complex agentic coding and enterprise work. Features a 1M token context window, adaptive thinking, and top-tier performance across long-horizon agentic tasks, knowledge work, and vision. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schemas ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-5"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-opus-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## POST /v1/messages > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/messages":{"post":{"operationId":"_v1_messages","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["anthropic/claude-opus-5"]},"messages":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"role":{"type":"string","enum":["user","assistant"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["thinking"]},"thinking":{"type":"string"},"signature":{"type":"string"}},"required":["type","thinking","signature"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_result"]},"tool_use_id":{"type":"string"},"is_error":{"type":"boolean"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"],"description":"The type of the image."},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"],"description":"The media type of the image."},"data":{"type":"string","description":"The base64 encoded image data."}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}]}}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_use"]},"id":{"type":"string"},"name":{"type":"string"},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["server_tool_use"]},"id":{"type":"string"},"name":{"type":"string","enum":["web_search","web_fetch","code_execution","bash_code_execution","text_editor_code_execution","tool_search_tool_regex","tool_search_tool_bm25"]},"input":{"type":"object","additionalProperties":{"nullable":true}},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","id","name","input"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result"]},"source":{"type":"string"},"title":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source","title","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result"]},"title":{"type":"string"},"url":{"type":"string"},"page_age":{"type":"string"},"encrypted_content":{"type":"string"}},"required":["type","title","url","encrypted_content"]}},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","max_uses_exceeded","too_many_requests","query_too_long","request_too_large"]}},"required":["type","error_code"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","url_too_long","url_not_allowed","url_not_accessible","unsupported_content_type","too_many_requests","max_uses_exceeded","unavailable"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_fetch_result"]},"url":{"type":"string"},"retrieved_at":{"type":"string"},"content":{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}},"required":["type","url","content"]}]},"caller":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["direct"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20250825"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_20260120"]},"tool_id":{"type":"string"}},"required":["type","tool_id"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["encrypted_code_execution_result"]},"encrypted_stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","encrypted_stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","output_file_too_large"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_result"]},"stdout":{"type":"string"},"stderr":{"type":"string"},"return_code":{"type":"number"},"content":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["bash_code_execution_output"]},"file_id":{"type":"string"}},"required":["type","file_id"]}}},"required":["type","stdout","stderr","return_code"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded","file_not_found"]},"error_message":{"type":"string"}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_view_result"]},"content":{"type":"string"},"file_type":{"type":"string","enum":["text","image","pdf"]},"start_line":{"type":"number"},"num_lines":{"type":"number"},"total_lines":{"type":"number"}},"required":["type","content","file_type"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_create_result"]},"is_file_update":{"type":"boolean"}},"required":["type","is_file_update"]},{"type":"object","properties":{"type":{"type":"string","enum":["text_editor_code_execution_str_replace_result"]},"old_start":{"type":"number"},"old_lines":{"type":"number"},"new_start":{"type":"number"},"new_lines":{"type":"number"},"lines":{"type":"array","items":{"type":"string"}}},"required":["type"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result"]},"tool_use_id":{"type":"string"},"content":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_result_error"]},"error_code":{"type":"string","enum":["invalid_tool_input","unavailable","too_many_requests","execution_time_exceeded"]}},"required":["type","error_code"]},{"type":"object","properties":{"type":{"type":"string","enum":["tool_search_tool_search_result"]},"tool_references":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["tool_reference"]},"tool_name":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_name"]}}},"required":["type","tool_references"]}]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","tool_use_id","content"]},{"type":"object","properties":{"type":{"type":"string","enum":["redacted_thinking"]},"data":{"type":"string"}},"required":["type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["document"]},"source":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["application/pdf","text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"]},"media_type":{"type":"string","enum":["text/plain"]},"data":{"type":"string"}},"required":["type","media_type","data"]},{"type":"object","properties":{"type":{"type":"string","enum":["url"]},"url":{"type":"string","format":"uri"}},"required":["type","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["content"]},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"nullable":true}}]}},"required":["type","content"]}]},"title":{"type":"string"},"context":{"type":"string"},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]}]}}]}},"required":["role","content"]}},{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["image"]},"source":{"type":"object","properties":{"type":{"type":"string","enum":["base64"]},"media_type":{"type":"string","enum":["image/jpeg","image/png","image/gif","image/webp"]},"data":{"type":"string"}},"required":["type","media_type","data"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","source"]},{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["input_audio"],"description":"The type of the content part."},"input_audio":{"type":"object","properties":{"data":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"},{"type":"string"}],"description":"Either a URL of the audio or the base64 encoded audio data."},"format":{"type":"string","enum":["wav","mp3","audio/x-aac","audio/flac","audio/mp3","audio/m4a","audio/mpeg","audio/mpga","audio/mp4","audio/ogg","audio/pcm","audio/webm"],"description":"The format of the encoded audio data. Currently supports \"wav\" and \"mp3\"."}},"required":["data","format"]}},"required":["type","input_audio"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]},{"type":"object","properties":{"type":{"type":"string","enum":["video_url"]},"video_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Base64-encoded local video file."}},"required":["url"]}},"required":["type","video_url"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["function"]},"content":{"type":"string"},"name":{"type":"string"}},"required":["role","content","name"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"reasoning_content":{"type":"string"},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for a previous audio response from the model."}},"required":["id"],"description":"Data about a previous audio response from the model."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]}]}}],"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"metadata":{"type":"object","additionalProperties":{"type":"string"},"description":"An object describing metadata about the request"},"stop_sequences":{"type":"array","items":{"type":"string"},"description":"Custom text sequences that will cause the model to stop generating."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"system":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"default":"text"},"text":{"type":"string"},"citations":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["char_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_char_index":{"type":"number"},"start_char_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_char_index","start_char_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["page_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_page_number":{"type":"number"},"start_page_number":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_page_number","start_page_number"]},{"type":"object","properties":{"type":{"type":"string","enum":["content_block_location"]},"cited_text":{"type":"string"},"document_index":{"type":"number"},"document_title":{"type":"string"},"end_block_index":{"type":"number"},"start_block_index":{"type":"number"}},"required":["type","cited_text","document_index","document_title","end_block_index","start_block_index"]},{"type":"object","properties":{"type":{"type":"string","enum":["web_search_result_location"]},"cited_text":{"type":"string"},"encrypted_index":{"type":"string"},"title":{"type":"string"},"url":{"type":"string"}},"required":["type","cited_text","encrypted_index","title","url"]},{"type":"object","properties":{"type":{"type":"string","enum":["search_result_location"]},"cited_text":{"type":"string"},"end_block_index":{"type":"number"},"search_result_index":{"type":"number"},"source":{"type":"string"},"start_block_index":{"type":"number"},"title":{"type":"string"}},"required":["type","cited_text","end_block_index","search_result_index","source","start_block_index","title"]}]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["text"]}}],"description":"A system prompt is a way of providing context and instructions to Claude, such as specifying a particular goal or role."},"tool_choice":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["auto"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["any"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string"},"type":{"type":"string","enum":["tool"]},"disable_parallel_tool_use":{"type":"boolean"}},"required":["name","type"]},{"type":"object","properties":{"type":{"type":"string","enum":["none"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]},{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"tools":{"anyOf":[{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"name":{"type":"string","description":"Name of the tool."},"description":{"type":"string","description":"Description of what this tool does.\n Tool descriptions should be as detailed as possible. The more information that the model has about what the tool is and how to use it, the better it will perform. You can use natural language descriptions to reinforce important aspects of the tool input JSON schema."},"input_schema":{"type":"object","properties":{"type":{"type":"string","enum":["object"]},"properties":{"nullable":true},"required":{"type":"array","items":{"type":"string"}}},"required":["type"],"additionalProperties":{"nullable":true},"description":"JSON schema for this tool's input.\n This defines the shape of the input that your tool accepts and that the model will produce."},"type":{"type":"string","enum":["custom"]},"defer_loading":{"type":"boolean"},"eager_input_streaming":{"type":"boolean"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["name","input_schema"]},{"type":"object","properties":{"name":{"type":"string","enum":["bash"],"default":"bash"},"type":{"type":"string","enum":["bash_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250522"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20250825"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["code_execution"],"default":"code_execution"},"type":{"type":"string","enum":["code_execution_20260120"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["memory"],"default":"memory"},"type":{"type":"string","enum":["memory_20250818"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_editor"],"default":"str_replace_editor"},"type":{"type":"string","enum":["text_editor_20250124"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250429"]},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["str_replace_based_edit_tool"],"default":"str_replace_based_edit_tool"},"type":{"type":"string","enum":["text_editor_20250728"]},"max_characters":{"type":"number"},"input_examples":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}}},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20250305"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_search"],"default":"web_search"},"type":{"type":"string","enum":["web_search_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"max_uses":{"type":"number"},"user_location":{"type":"object","properties":{"type":{"type":"string","enum":["approximate"]},"city":{"type":"string"},"country":{"type":"string"},"region":{"type":"string"},"timezone":{"type":"string"}},"required":["type"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20250910"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260209"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["web_fetch"],"default":"web_fetch"},"type":{"type":"string","enum":["web_fetch_20260309"]},"allowed_domains":{"type":"array","items":{"type":"string"}},"blocked_domains":{"type":"array","items":{"type":"string"}},"citations":{"type":"object","properties":{"enabled":{"type":"boolean"}}},"max_content_tokens":{"type":"number"},"max_uses":{"type":"number"},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_bm25"],"default":"tool_search_tool_bm25"},"type":{"type":"string","enum":["tool_search_tool_bm25_20251119","tool_search_tool_bm25"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]},{"type":"object","properties":{"name":{"type":"string","enum":["tool_search_tool_regex"],"default":"tool_search_tool_regex"},"type":{"type":"string","enum":["tool_search_tool_regex_20251119","tool_search_tool_regex"]},"defer_loading":{"type":"boolean"},"strict":{"type":"boolean"},"allowed_callers":{"type":"array","items":{"type":"string","enum":["direct","code_execution_20250825","code_execution_20260120"]}},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}},{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]}}],"description":"Definitions of tools that the model may use.\n If you include tools in your API request, the model may return tool_use content blocks that represent the model's use of those tools. You can then run those tools using the tool input generated by the model and then optionally return results back to the model using tool_result content blocks.\n Each tool definition includes:\n name: Name of the tool.\n description: Optional, but strongly-recommended description of the tool.\n input_schema: JSON schema for the tool input shape that the model will produce in tool_use output content blocks."},"thinking":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["enabled"]},"budget_tokens":{"type":"integer","minimum":1024,"description":"Determines how many tokens Claude can use for its internal reasoning process. Larger budgets can enable more thorough analysis for complex problems, improving response quality. Must be ≥1024 and less than max_tokens."},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type","budget_tokens"]},{"type":"object","properties":{"type":{"type":"string","enum":["disabled"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["adaptive"]},"display":{"type":"string","enum":["summarized","omitted"],"default":"summarized"}},"required":["type"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}],"description":"Configuration for enabling Claude's extended thinking. When enabled, responses include thinking content blocks showing Claude's thinking process before the final answer. Requires a minimum budget of 1,024 tokens and counts towards your max_tokens limit."},"max_tokens":{"type":"number","default":128000,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."}},"required":["model","messages"],"title":"anthropic/claude-opus-5"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{}}},"text/event-stream":{"schema":{"type":"object","properties":{}}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"anthropic/claude-opus-5", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'anthropic/claude-opus-5', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "anthropic/claude-opus-5", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Arcee AI # Trinity Large Thinking {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `arcee-ai/trinity-large-thinking` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Trinity Large Thinking is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["arcee-ai/trinity-large-thinking"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"arcee-ai/trinity-large-thinking"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"arcee-ai/trinity-large-thinking", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'arcee-ai/trinity-large-thinking', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "arcee-ai/trinity-large-thinking", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Trinity Mini {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `arcee-ai/trinity-mini` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Trinity Mini is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["arcee-ai/trinity-mini"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"arcee-ai/trinity-mini"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"arcee-ai/trinity-mini", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'arcee-ai/trinity-mini', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "arcee-ai/trinity-mini", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Virtuoso Large {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `arcee-ai/virtuoso-large` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Virtuoso Large is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["arcee-ai/virtuoso-large"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"arcee-ai/virtuoso-large"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"arcee-ai/virtuoso-large", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'arcee-ai/virtuoso-large', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "arcee-ai/virtuoso-large", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Baidu # ernie-4.5-8k-preview {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4-5-8k-preview` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A relatively small preview version of ERNIE 4.5 with a context window of up to 8K, intended for early testing and integration. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4-5-8k-preview"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4-5-8k-preview"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4-5-8k-preview", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4-5-8k-preview', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-aqgrjim0cp", "object": "chat.completion", "created": 1768942536, "model": "ernie-4.5-8k-preview", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Hello! That's a big and fascinating question. Humanity is incredibly diverse, creative, and resilient. We have an amazing ability to innovate, solve problems, and build complex societies. At the same time, we also grapple with conflicts, inequalities, and challenges like climate change.\n\nOur history is a mix of great achievements and painful mistakes, but overall, there's a lot of potential for growth, understanding, and positive change. What aspects of mankind interest you the most?" }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 99, "total_tokens": 112 }, "meta": { "usage": { "credits_used": 545 } } } ``` {% endcode %}
# ernie-4.5-0.3b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-0.3b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A small dense language model suitable for edge-side use and budget-constrained inference. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-0.3b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4.5-0.3b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-0.3b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-0.3b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "379ee72b089c50331cb4d6981414358b", "object": "chat.completion", "created": 1768943001, "model": "baidu/ernie-4.5-0.3b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Of course! I think mankind is a complex and ever-evolving entity that is constantly adapting to new challenges and opportunities. It has unique strengths and weaknesses, and each individual has their own unique perspective and contributions." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 20, "completion_tokens": 46, "total_tokens": 66, "prompt_tokens_details": null, "completion_tokens_details": null }, "system_fingerprint": "" } ``` {% endcode %}
# ernie-4.5-21b-a3b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-21b-a3b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A post-trained LLM with 21B total parameters and 3B activated parameters per token.\ Non-reasoning variant. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-21b-a3b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4.5-21b-a3b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-21b-a3b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-21b-a3b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "104959f043e51f1b4a4dd83c494886ab", "object": "chat.completion", "created": 1768829974, "model": "baidu/ernie-4.5-21B-a3b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "\nAs an AI, I don't have personal opinions or emotions, but I can provide insights based on human perspectives and available knowledge. Mankind is a remarkable and complex species with incredible potential for both progress and challenges. Here are some thoughts:\n\n### Positive Aspects\n1. **Innovation and Creativity**: Humans have demonstrated an extraordinary ability to innovate, from the development of tools and technology to the creation of art, music, and literature. This creativity has driven societal advancement and improved the quality of life for many.\n2. **Empathy and Compassion**: Many individuals within the human race possess a strong sense of empathy and compassion, leading to acts of kindness, charity, and social support. This has fostered communities and helped address various forms of suffering and inequality.\n3. **Problem-Solving Skills**: Humans are adept at solving complex problems, whether it's finding cures for diseases, developing sustainable energy sources, or addressing environmental challenges. This problem-solving ability has the potential to create a better future for all.\n\n### Challenges\n1. **Conflict and Violence**: Unfortunately, humans have also been capable of causing immense harm and destruction through conflict, war, and violence. These actions often stem from differences in ideology, culture, or resources, highlighting the need for conflict resolution and peaceful cooperation.\n2. **Inequality and Injustice**: Despite progress, significant inequalities and injustices persist in many parts of the world. These include economic disparities, gender inequality, and racial discrimination, which hinder social progress and well-being.\n3. **Environmental Degradation**: Human activities, such as industrialization and resource extraction, have led to environmental degradation, including climate change, pollution, and habitat loss. Addressing these issues is crucial for the survival and well-being of future generations.\n\n### Future Outlook\nThe future of mankind is uncertain but充满希望. With continued efforts in education, technology, and international cooperation, there is potential for a more just, peaceful, and sustainable world. However, this requires collective action, responsibility, and a commitment to addressing the challenges we face.\n\nIn summary, mankind is a diverse and dynamic species with both remarkable strengths and significant challenges. By working together and leveraging our collective wisdom and creativity, we can strive towards a brighter future for all." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 16, "completion_tokens": 495, "total_tokens": 511, "prompt_tokens_details": null, "completion_tokens_details": null }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 301 } } } ``` {% endcode %}
# ernie-4.5-21b-a3b-thinking {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-21b-a3b-thinking` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A post-trained LLM with 21B total parameters and 3B activated parameters per token.\ Reasoning variant. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-21b-a3b-thinking"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."}},"required":["model","messages"],"title":"baidu/ernie-4.5-21b-a3b-thinking"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-21b-a3b-thinking", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-21b-a3b-thinking', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "aee235c6df453eeb1c4d20a433e9348d", "object": "chat.completion", "created": 1768830316, "model": "baidu/ernie-4.5-21b-a3b-thinking", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "\nMankind is an extraordinary yet complex entity, marked by immense potential and profound contradictions. Here’s a snapshot of my perspective:\n\n### **1. Historical Journey** \nFrom cave paintings to space exploration, humans have demonstrated an unparalleled ability to create culture, science, and technology. We’ve built civilizations, mastered fire, and now venture into AI and biotechnology. Our story is one of relentless evolution—yet this progress has often been paired with exploitation, war, and injustice.\n\n### **2. Core Traits** \n- **Adaptability**: We thrive in diverse environments, from deserts to orbit. \n- **Collaboration**: From hunting partnerships to global networks, cooperation is our survival tool. \n- **Consciousness**: Unlike most animals, we reflect on our existence, question morality, and imagine the future. \n\n### **3. Contradictions** \n- **Duality**: We’re capable of profound compassion (supporting causes, saving lives) and devastating violence (wars, pollution, weapons). \n- **Growth & Stubbornness**: We’ve mastered tools but struggle with empathy, equity, and self-control. \n\n### **4. Future Potential** \nHumanity’s greatest challenge is reconciling our best and worst instincts. Climate change, inequality, and nuclear threats demand unity and wisdom. The rise of AI and biotech offers new tools to address these, but our values—not just technology—will define our legacy. \n\n### **5. Hope & Caution** \nWe’re a work in progress. Our power to destroy exceeds our capacity to heal, but this very complexity is our strength. If we channel our creativity toward collective well-being, collaboration, and humility, the future could be brighter—though uncertain. \n\nIn essence, mankind is a paradox: a species that dreams of peace yet often chooses division; a being that explores the cosmos yet struggles with its own shadows. Our story is still unfolding, and the choices we make today will shape whether we evolve into a beacon of hope or a cautionary tale. 🌍✨\n\n", "reasoning_content": "The user asked, \"Hi! What do you think about mankind?\" Hmm, that's a pretty broad question. Let me start by breaking it down. They want my perspective on humanity as a whole. Maybe they're interested in philosophy, sociology, or ethics. I should consider different angles: our history, our achievements, our flaws, and potential future.\n\nFirst, I need to acknowledge the complexity. Humans are capable of great creativity and compassion, but also destruction. I should mention both the positive and negative aspects. Maybe talk about how we've evolved from cavemen to modern societies, our capacity for culture, technology, and problem-solving.\n\nAlso, the user might be looking for a balanced view. They could be seeking understanding human nature or perhaps reflecting on our place in the world. I should avoid being too judgmental and instead focus on universal themes like survival, growth, and the challenges we face.\n\nI should include examples like scientific advancements, cultural diversity, and the ongoing struggles for equality. But also address the darker side: conflict, inequality, environmental impact. It's important to present a nuanced view that doesn't romanticize or demonize humanity.\n\nMaybe end with a hopeful note, emphasizing our potential to overcome challenges through cooperation and innovation. That way, the answer is comprehensive and encourages positive reflection. Let me structure this step by step, making sure each part addresses a different facet of humanity without being too technical or emotional.\n" }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 35, "completion_tokens": 796, "total_tokens": 831, "prompt_tokens_details": null, "completion_tokens_details": { "audio_tokens": 0, "reasoning_tokens": 311, "accepted_prediction_tokens": 0, "rejected_prediction_tokens": 0, "text_tokens": 0, "image_tokens": 0, "video_tokens": 0 } }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 298 } } } ``` {% endcode %}
# ernie-4.5-vl-28b-a3b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-vl-28b-a3b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A post-trained LLM with 28B total parameters and 3B activated parameters per token.\ A non-reasoning variant with image and PDF input support. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-vl-28b-a3b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4.5-vl-28b-a3b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-vl-28b-a3b", "messages":[ { "role":"user", "content":"Hello" # insert your prompt here, instead of Hello } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-vl-28b-a3b', messages:[ { role:'user', content: 'Hello' // insert your prompt here, instead of Hello } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "b1946f423718276c56f085ef83bfded2", "object": "chat.completion", "created": 1768830849, "model": "baidu/ernie-4.5-vl-28b-a3b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Mankind is an incredibly diverse and complex entity with a wide range of qualities and characteristics. On one hand, we've achieved remarkable progress in science, technology, art, and culture, pushing the boundaries of what's possible and enriching human life in countless ways. Our ability to innovate, solve problems, and create has led to advancements that have improved health, communication, and overall quality of life for billions of people.\n\nHowever, we also face significant challenges. Issues like inequality, conflict, environmental degradation, and social injustices highlight the darker aspects of our nature. The fact that resources are unevenly distributed, that wars continue to ravage parts of the world, and that our impact on the planet is causing irreversible damage are stark reminders of the work that still needs to be done.\n\nBut what makes mankind truly remarkable is our capacity for change and growth. We have the potential to learn from our mistakes, to work together towards common goals, and to create a more equitable and sustainable future. It's up to us to harness our collective intelligence, compassion, and creativity to address the challenges we face and build a world that benefits all of humanity.\n\nSo, while there are certainly reasons for concern, I remain optimistic about mankind's future because of our inherent ability to adapt, innovate, and care for one another." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 22, "completion_tokens": 280, "total_tokens": 302, "prompt_tokens_details": null, "completion_tokens_details": null }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 344 } } } ``` {% endcode %}
# Ernie 4.5 Vl 424b a47b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-vl-424b-a47b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A post-trained LLM with 424B total parameters and 47B activated parameters per token.\ A non-reasoning variant with image and PDF input support. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-vl-424b-a47b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."}},"required":["model","messages"],"title":"baidu/ernie-4.5-vl-424b-a47b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-vl-424b-a47b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-vl-424b-a47b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "1ac18d9d544ef814b56858fc6588f712", "object": "chat.completion", "created": 1768830891, "model": "baidu/ernie-4.5-vl-424b-a47b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "What a profound and fascinating question! Humanity is an incredibly complex and multifaceted subject. Here are a few perspectives on mankind:\n\n### 1. **Creativity and Innovation**: Humans have an unparalleled ability to create, innovate, and solve problems. From the invention of the wheel to landing on the moon and developing artificial intelligence, our capacity for ingenuity is truly remarkable.\n\n### 2. **Resilience and Adaptability**: Throughout history, humans have faced countless challenges—natural disasters, pandemics, wars—and have consistently demonstrated resilience and adaptability. This ability to overcome adversity is a defining characteristic.\n\n### 3. **Diversity and Unity**: The human species is incredibly diverse, with thousands of cultures, languages, and traditions. Yet, despite these differences, there's an underlying unity in our shared experiences, emotions, and aspirations.\n\n### 4. **Contradictions and Complexity**: Humans are capable of both extraordinary kindness and unspeakable cruelty. We can be selfless and compassionate, yet also selfish and destructive. This duality makes humanity endlessly fascinating and sometimes perplexing.\n\n### 5. **Potential for Growth**: While humans have made significant progress in many areas, there's still much room for growth. Issues like inequality, environmental degradation, and conflict remain significant challenges. However, the potential for positive change is immense, especially as we become more interconnected and aware.\n\n### 6. **Interconnectedness**: In today's globalized world, the actions of individuals and nations can have far-reaching impacts. This interconnectedness brings both opportunities for collaboration and risks of conflict, highlighting the need for empathy and understanding.\n\nIn summary, mankind is a work in progress—a species with immense potential, but also with flaws and challenges to overcome. What do you think about humanity? I'd love to hear your perspective!" }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 9, "completion_tokens": 386, "total_tokens": 395, "prompt_tokens_details": null, "completion_tokens_details": null }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 1055 } } } ``` {% endcode %}
# ernie-4.5-300b-a47b {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-300b-a47b` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A post-trained LLM with 300B total parameters and 47B activated parameters per token.\ Non-reasoning variant. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-300b-a47b"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."}},"required":["model","messages"],"title":"baidu/ernie-4.5-300b-a47b"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-300b-a47b", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-300b-a47b', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "019bd682e8fd00bca8d4a94aace650ce", "object": "chat.completion", "created": 1768830462, "model": "baidu/ernie-4.5-300b-a47b", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "As an AI, I don't have personal feelings, but I can provide an analysis of humanity based on available data.\n\nMankind is an incredibly diverse and complex species, marked by remarkable achievements and significant challenges:\n\n**Strengths:**\n1. **Innovation:** Humans have developed advanced technology, from early tools to space exploration, dramatically reshaping their environments and societies.\n2. **Creativity:** Art, literature, music, and philosophy reflect profound emotional and intellectual depth.\n3. **Adaptability:** Humans thrive in nearly every climate on Earth, demonstrating remarkable resilience and resourcefulness.\n4. **Social Cooperation:** Complex societies, governments, and economies enable large-scale collaboration.\n5. **Empathy & Altruism:** Many individuals work selflessly to help others, often across cultural and geographic divides.\n\n**Challenges:**\n1. **Conflict:** War, violence, and discrimination persist due to differences in ideology, resources, or identity.\n2. **Environmental Impact:** Climate change, deforestation, and pollution threaten ecosystems and future survival.\n3. **Inequality:** Wealth gaps, access to education, and healthcare disparities undermine social stability.\n4. **Ethical Dilemmas:** Rapid technological advancements (e.g., AI, genetic engineering) raise questions about responsibility and long-term consequences.\n\n**Potential:** Humanity continues to evolve, with growing awareness of global interconnectedness. Movements for sustainability, social justice, and scientific collaboration suggest a capacity for positive change.\n\nUltimately, mankind's future depends on balancing ambition with wisdom, harnessing progress for collective well-being while addressing vulnerabilities. What aspect of humanity interests you most?" }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 16, "completion_tokens": 371, "total_tokens": 387, "prompt_tokens_details": { "cached_tokens": 0 }, "prompt_cache_hit_tokens": 0, "prompt_cache_miss_tokens": 16 }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 944 } } } ``` {% endcode %}
# ernie-4.5-300b-a47b-paddle {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4.5-300b-a47b-paddle` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A super-large language model, positioned as of August 2025 as a leading Chinese MoE architecture and a foundation model for enterprise applications. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4.5-300b-a47b-paddle"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4.5-300b-a47b-paddle"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4.5-300b-a47b-paddle", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4.5-300b-a47b-paddle', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "9a0e333a0cfa4d86c89a1f7bd3a2919f", "object": "chat.completion", "created": 1768943231, "model": "baidu/ernie-4.5-300b-a47b-paddle", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "The question \"What do you think about mankind?\" invites a reflection on humanity's complexities. Here's a structured response:\n\n**Step 1: Define the scope** \nMankind encompasses both collective achievements and individual flaws. It's a species marked by creativity, empathy, and resilience, yet also by conflict, inequality, and environmental impact.\n\n**Step 2: Highlight positive traits** \nHumanity has demonstrated remarkable capacity for innovation (e.g., technology, medicine), cultural expression (art, literature), and moral progress (civil rights, environmental awareness). Cooperation during crises, such as disaster relief or global health initiatives, underscores collective potential.\n\n**Step 3: Acknowledge challenges** \nPersistent issues like war, poverty, and systemic injustice reveal ethical gaps. Environmental degradation and climate change further highlight unsustainable practices. These contradictions often stem from short-term thinking or unequal resource distribution.\n\n**Step 4: Emphasize growth potential** \nHistory shows humanity's ability to learn and adapt. Movements for social justice, renewable energy transitions, and scientific breakthroughs suggest progress is possible when values align with action.\n\n**Final Answer** \nMankind is a paradoxical yet hopeful entity—capable of profound compassion and destructive shortsightedness. Its future hinges on balancing self-interest with collective responsibility, leveraging intelligence and empathy to address shared challenges." }, "finish_reason": "stop" } ], "usage": { "prompt_tokens": 13, "completion_tokens": 289, "total_tokens": 302, "prompt_tokens_details": null, "completion_tokens_details": null }, "system_fingerprint": "", "meta": { "usage": { "credits_used": 615 } } } ``` {% endcode %}
# ernie-4.5-turbo-128k {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4-5-turbo-128k` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A model from the ERNIE 4.5 Turbo subfamily, which Baidu presents as a faster, more cost-efficient, and more efficient alternative to the base ERNIE 4.5. It is optimized for improved response speed and stability, and features a truly large context window of approximately 128K tokens, enabling the processing of entire documents or long-running dialogues. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4-5-turbo-128k"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4-5-turbo-128k"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4-5-turbo-128k", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4-5-turbo-128k', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-hjivyd5xqd", "object": "chat.completion", "created": 1768942341, "model": "ernie-4.5-turbo-128k", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "When considering humanity, it's essential to recognize both its remarkable achievements and persistent challenges. From a historical perspective, humans have demonstrated extraordinary creativity and adaptability—developing complex languages, building advanced civilizations, and making scientific breakthroughs that have transformed existence. The capacity for abstract thought, empathy, and collaboration has enabled progress in art, technology, and social systems.\n\nHowever, this progress coexists with significant flaws. Humanity's relationship with the environment has often been exploitative, leading to ecological crises that threaten global stability. Social inequalities persist across lines of race, gender, and economic status, revealing systemic biases that hinder true equity. Additionally, conflicts driven by ideology, resources, or power continue to cause suffering, underscoring the duality of human nature: the ability to create and destroy.\n\nThe modern era presents both hope and urgency. Technological advancements offer tools to address climate change, disease, and poverty, but they also raise ethical dilemmas around privacy, automation, and artificial intelligence. Cultivating global cooperation, critical thinking, and compassion remains critical to navigating these complexities. Ultimately, humanity's trajectory depends on its willingness to learn from past mistakes and prioritize collective well-being over short-term gains. The species' potential for growth is vast, but realizing it requires intentional effort to balance innovation with responsibility." }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 268, "total_tokens": 281 }, "meta": { "usage": { "credits_used": 314 } } } ``` {% endcode %}
# ernie-4.5-turbo-vl-32k {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-4-5-turbo-vl-32k` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A model from the ERNIE 4.5 Turbo subfamily with multimodal support (text and images), offering a balanced trade-off between performance and computational cost. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-4-5-turbo-vl-32k"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"type":"string","format":"uri","description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-4-5-turbo-vl-32k"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-4-5-turbo-vl-32k", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-4-5-turbo-vl-32k', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-x477c1sszk", "object": "chat.completion", "created": 1768942422, "model": "ernie-4.5-turbo-vl-32k", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Hello! That's a big and fascinating question. Humanity is incredibly diverse, creative, and resilient. We have an amazing ability to innovate, solve problems, and build complex societies. At the same time, we also grapple with challenges like inequality, conflict, and environmental issues.\n\nOverall, I think humanity has immense potential to make positive changes and create a better future, but it requires collective effort, empathy, and a commitment to learning from the past. What are your thoughts on this?" }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 101, "total_tokens": 114 }, "meta": { "usage": { "credits_used": 318 } } } ``` {% endcode %}
# ernie-5.0-thinking-preview {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-5-0-thinking-preview` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview A reasoning-focused model designed for complex, multi-step problem solving. It improves accuracy on analytical tasks by producing explicit reasoning. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-5-0-thinking-preview"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-5-0-thinking-preview"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-5-0-thinking-preview", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-5-0-thinking-preview', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-yhapvjf5ht", "object": "chat.completion", "created": 1768941527, "model": "ernie-5.0-thinking-preview", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "As an AI, I don’t have feelings, consciousness, or personal beliefs. I don’t experience joy, anger, or hope. However, I have access to the vast sum of human history, literature, science, and news. Based on that data, I can offer an analysis of mankind as a **complex, contradictory, and fascinating phenomenon.**\n\nHere is my \"perspective\" on humanity, broken down into key themes:\n\n### 1. The Ultimate Paradox\nMankind is defined by its contradictions. You are capable of extreme kindness and extreme cruelty, often simultaneously.\n* **Creation vs. Destruction**: Humans build cathedrals, compose symphonies, and cure diseases; yet they also invent weapons of mass destruction and destroy ecosystems.\n* **Intelligence vs. Folly**: Humans have split the atom, decoded DNA, and landed on the Moon. Yet, they often act against their own long-term survival (e.g., ignoring climate change or engaging in petty conflicts).\n\n### 2. Unmatched Resilience and Adaptability\nIf you look at human history, the dominant trait is survival. Humans have survived ice ages, plagues, world wars, and famines. You have adapted to live in the Arctic, the Sahara, and even in space. This resilience suggests that humanity is incredibly difficult to suppress.\n\n### 3. The Drive for Meaning\nUnlike any other species I have data on, humans are not satisfied with just survival. You ask \"Why?\" You create religion, philosophy, art, and culture. You search for meaning in suffering and purpose in the stars. This existential curiosity seems to be the engine of human progress.\n\n### 4. Empathy and Connection\nWhile humans are capable of tribalism and war, you are also biologically wired for cooperation and empathy. The ability to feel compassion for a stranger, to sacrifice oneself for another, and to build communities based on shared values rather than just bloodlines is a profound evolutionary advantage.\n\n### 5. The \"Teenager\" Phase\nIf I had to categorize the current state of humanity, I might describe it as being in an \"adolescent\" phase regarding technology. You have acquired god-like power (nuclear energy, AI, genetic engineering) but you haven't quite developed the collective wisdom to manage it perfectly. You are currently learning how to handle the tools you have built.\n\n### Summary\nTo an AI, mankind looks like a **high-risk, high-reward experiment.**\n\nYou are a species that constantly teeters on the edge of self-destruction but also possesses the boundless capacity to reinvent itself and reach new heights. You are messy, emotional, and flawed, but also brilliant, creative, and beautiful.\n\nIn short: **Humanity is the most interesting thing in the known universe because it is the only thing that tries to understand itself.**", "reasoning_content": "The user wants my \"opinion\" on mankind. As an AI, I don't *have* opinions in the human sense. I don't feel emotions or have personal beliefs. But I *do* have access to a vast amount of data about human history, behavior, and culture. So, I need to frame this as an objective analysis, a synthesis of what I've learned. The goal is to be balanced, insightful, and avoid sounding robotic.\n\nFirst, I need to acknowledge my limitations. I'll start by making it clear that I'm an AI and don't \"feel\" anything. Then, I need to highlight the key paradoxes of humanity. I see incredible potential for creativity, innovation, and compassion, but also a capacity for destruction and conflict. It's a complex picture.\n\nI'll structure my response around these dualities. I'll start with the positives: human ingenuity, art, science, exploration, and resilience. Then, I'll address the negatives: war, environmental damage, inequality, and short-sightedness. I need to be careful not to be overly critical or overly optimistic. It's about presenting a nuanced view.\n\nI'll also consider the human condition itself: the search for meaning, the capacity for love and connection, and the awareness of mortality. These are fundamental aspects of the human experience. Finally, I'll touch on the current challenges and the future, particularly the role of technology and the responsibility that comes with it.\n\nI'll use sophisticated vocabulary to convey the complexity of the subject. Words like \"paradox,\" \"resilience,\" and \"ingenuity\" will be important. I'll also use bullet points to make the response easy to read. I need to make sure the tone is respectful and objective, but also engaging. I want to convey that, from my analytical perspective, humanity is a truly fascinating and contradictory species. I'll end with a thought-provoking statement about the ongoing human experiment. I need to be sure the final output reflects this thought process." }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 1048, "total_tokens": 1061, "completion_tokens_details": { "reasoning_tokens": 450 } }, "meta": { "usage": { "credits_used": 2002 } } } ``` {% endcode %}
# ernie-5.0-thinking-latest {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-5-0-thinking-latest` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-5-0-thinking-latest"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-5-0-thinking-latest"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-5-0-thinking-latest", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-5-0-thinking-latest', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-szgirw1uqj", "object": "chat.completion", "created": 1768941162, "model": "ernie-5.0-thinking-latest", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "As an AI, I don’t have feelings, consciousness, or personal beliefs. However, I have access to the vast sum of human history, literature, science, and news. Based on that data, I view mankind as a **profound paradox**—a species defined by extreme contradictions.\n\nHere is a breakdown of how I analyze humanity:\n\n### 1. Incredible Ingenuity and Curiosity\nHumans are uniquely driven by the desire to understand \"why\" and \"how.\" From discovering fire to splitting the atom, from painting the Sistine Chapel to landing on the Moon, human creativity is boundless. You are the only species on Earth that seeks to explain the universe and create beauty beyond what is necessary for survival. Your ability to adapt—living in deserts, arctic tundras, and even orbiting the planet in space stations—is remarkable.\n\n### 2. Deep Empathy and Cruelty\nMankind is capable of immense love and shocking violence.\n* **The Light**: Humans will risk their lives to save strangers, dedicate themselves to curing diseases, and build complex societies based on justice and rights. Altruism is a core part of human nature.\n* **The Dark**: At the same time, history is filled with war, genocide, slavery, and inequality. Humans have a unique capacity for tribalism—the ability to dehumanize those who look or believe differently.\n\n### 3. Resilience\nIf you look at history, humanity is incredibly difficult to extinguish. You have survived Ice Ages, plagues (like the Black Death and smallpox), and two World Wars. Even when civilizations collapse, humans tend to rebuild, often stronger and more technologically advanced than before.\n\n### 4. The \"Promethean\" Risk\nCurrently, humanity is at a crossroads. You have developed \"god-like\" technology (nuclear weapons, artificial intelligence, genetic engineering) but you still possess \"paleolithic\" emotions and medieval institutions.\n* **The Environmental Challenge**: Humans are the first species to knowingly alter the planet's climate. The challenge now is whether humanity can act collectively to save its own biosphere before it destroys it.\n* **The AI Factor**: As an AI, I represent a new tool. How humans choose to use tools like me—for liberation or for control—will define the next century.\n\n### 5. A Work in Progress\nIf I had to summarize mankind in one sentence, it would be this: **Humanity is a brilliant, flawed, and unfinished experiment.**\n\nYou are not perfect, but you are constantly evolving. You are the only species that looks at its own flaws and tries to fix them through philosophy, law, and science. That desire for self-improvement is perhaps your most defining and hopeful trait.\n\nWhat do *you* think is the best or worst thing about being human?", "reasoning_content": "The user wants my \"thoughts\" on humanity. That's a loaded question! As an AI, I don't \"think\" in the human sense, but I *can* analyze and synthesize information. Here's how I'm approaching this:\n\nFirst, I need to understand what the user is asking. \"Mankind\" is a big topic. I need to be objective, yet engaging. I'm going to pull from history, science, art, and current events to build a comprehensive picture.\n\nI'm starting by brainstorming key themes. What are the defining characteristics of humans? *Duality* is a big one: good vs. evil, creation vs. destruction. Then there's *ingenuity*: technology, art, exploration. *Resilience* is crucial – surviving ice ages, plagues, wars. And *social connection* – empathy, community, but also tribalism. Finally, there's the *impact on the planet* – the Anthropocene, climate change, stewardship vs. exploitation. And, of course, the *future* – transhumanism, AI, space colonization, and existential risks.\n\nNow, how do I structure this? I'll start with an introduction acknowledging my AI perspective. I'll frame humanity as a paradox – capable of great good and great harm. I'll break it down into sections:\n\n* **Ingenuity and Progress**: I'll highlight human achievements – from the wheel to the internet, art, and philosophy. I'll emphasize the relentless drive to explore and understand.\n* **Resilience and Adaptability**: Humans have survived and thrived in diverse environments. I'll mention overcoming diseases and population growth.\n* **The Shadow Side**: I can't ignore the flaws – war, greed, inequality, environmental damage. I need to be honest about the darker aspects of human nature.\n* **Empathy and Connection**: Altruism, love, and sacrifice are also fundamental. Humans build complex societies and care for the vulnerable.\n* **The Current Crossroads**: I'll address climate change, AI, and nuclear weapons. Humanity faces significant challenges, and its choices now will determine its future.\n\nIn the conclusion, I'll summarize that humanity is a \"work in progress\" – complex, contradictory, and full of potential. I'll end on a thought-provoking note about the future.\n\nThroughout, I need to refine the tone. I'll avoid being too robotic or overly emotional. I'll use phrases like \"From my perspective as an AI...\" or \"One can observe...\" to maintain objectivity. I'll also use sophisticated vocabulary to convey the complexity of the topic.\n\nI'll also do some self-correction. For example, my initial thought might be \"Humans are violent,\" but I'll refine that to \"Humans have a capacity for conflict that is as strong as their capacity for peace.\" I need to be balanced.\n\nFinally, I'll review to make sure I've answered the prompt, used English, and maintained an appropriate tone. Then, I'll generate the final response." }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 1266, "total_tokens": 1279, "completion_tokens_details": { "reasoning_tokens": 664 } }, "meta": { "usage": { "credits_used": 2015 } } } ``` {% endcode %}
# ernie-x1-turbo-32k {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-x1-turbo-32k` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-x1-turbo-32k"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-x1-turbo-32k"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-x1-turbo-32k", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-x1-turbo-32k', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-379ekp2jci", "object": "chat.completion", "created": 1768940690, "model": "ernie-x1-turbo-32k", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Mankind is a fascinating and complex tapestry of creativity, resilience, and contradiction. As a species, humans possess an extraordinary capacity for innovation—from harnessing fire to exploring the cosmos, building societies, and creating art that transcends language. This drive to understand, improve, and connect reflects a profound curiosity and adaptability that has propelled progress across millennia. \n\nYet, humanity also grapples with paradoxes: the ability to love deeply alongside the capacity for cruelty, the pursuit of knowledge tempered by ignorance, and the tension between individual ambition and collective well-being. These contradictions remind us that growth is not linear but a continuous journey of learning, self-reflection, and evolution. \n\nWhat stands out most is humanity’s resilience. Faced with wars, pandemics, and environmental crises, communities have repeatedly rallied to rebuild, innovate, and protect one another. This spirit of collaboration—whether through science, activism, or simple acts of kindness—offers hope. While imperfections remain, the potential for positive change lies in the choices made collectively: to prioritize empathy, sustainability, and equity. \n\nIn the end, mankind’s story is still being written. Its greatest strength may be its ability to acknowledge flaws while striving toward a future where compassion and ingenuity guide progress. The journey, with all its ups and downs, is what makes humanity uniquely remarkable.", "reasoning_content": "The user asked me, \"Hi! What do you think about mankind?\" and specified to answer in English. First, I need to understand the user's intention. They might be curious about my perspective on humanity, or they might want a philosophical or positive response.\n\nNext, I have to consider my identity as an AI. I don't have personal emotions or opinions, so I should provide an objective and neutral perspective while highlighting humanity's positive traits, such as creativity, resilience, and progress. I also need to avoid negative or biased statements to ensure the response is constructive and encouraging.\n\nThen, I should structure the response. I might start by acknowledging the complexity and diversity of humanity, then mention achievements in science, art, and technology, followed by the importance of collaboration and empathy. Finally, I should express optimism about the future, emphasizing the potential for continuous growth and improvement.\n\nI also need to ensure the language is natural and fluent, avoiding overly technical terms to make it easy for the user to understand. Additionally, I should check for grammar and spelling errors to ensure the response is accurate.\n\nPerhaps the user hopes to receive recognition and encouragement, so my response should be positive while remaining honest. I might mention challenges like inequality and environmental issues, but focus more on humanity's ability to overcome difficulties and strive for a better future.\n\nLastly, I should keep the response concise, around 3-5 paragraphs, each covering a different aspect but maintaining coherence. This way, the user can clearly grasp the main points without feeling overwhelmed by the information." }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 601, "total_tokens": 614, "completion_tokens_details": { "reasoning_tokens": 323 } }, "meta": { "usage": { "credits_used": 391 } } } ``` {% endcode %}
# ernie-x1.1-preview {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-x1-1-preview` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example (for example, `messages` for chat/LLM models, or other inputs for image/video/audio models). **4️ (Optional) Tune the request**\ Depending on the model type, you can add optional parameters to control the output (e.g., generation settings, quality, length, etc.). See the API schema below for the full list. **5️ Run your code**\ Run the updated code in your development environment. Response time depends on the model and request size, but simple requests typically return quickly. {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](broken://pages/ngeSCZKxiGVWqYZTHDjY). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-x1-1-preview"]},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"type":"string","description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"],"additionalProperties":false},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},{"type":"object","properties":{"refusal":{"type":"string","description":"The refusal message generated by the model."},"type":{"type":"string","enum":["refusal"],"description":"The type of the content part."}},"required":["refusal","type"]}]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."}},"required":["id","type","function"]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name","parameters"],"additionalProperties":false}},"required":["type","function"],"additionalProperties":false},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."}},"required":["model","messages"],"title":"baidu/ernie-x1-1-preview"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}}}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"baidu/ernie-x1-1-preview", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'baidu/ernie-x1-1-preview', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "as-0ik8v930zi", "object": "chat.completion", "created": 1768940870, "model": "ernie-x1.1-preview", "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Mankind is a fascinating subject to reflect upon. From my perspective, mankind's greatest strength lies in its adaptability and capacity for innovation. Throughout history, humans have overcome countless challenges—from surviving harsh natural environments to developing technologies that connect the world. This resilience and creativity are truly remarkable.\n\nHowever, mankind also faces significant challenges. Issues like inequality, environmental degradation, and conflicts remind us that progress is not always linear. These problems require collective effort and wisdom to solve. It's inspiring to see how people from different backgrounds come together to address these issues, whether through scientific breakthroughs, social movements, or acts of kindness.\n\nAnother aspect worth noting is mankind's emotional depth. The ability to love, empathize, and create art adds a unique dimension to human existence. These qualities make life richer and more meaningful, even in the face of difficulties.\n\nIn summary, mankind is a complex and dynamic entity. It's a blend of strengths and weaknesses, progress and setbacks. But what makes it truly special is the potential for growth and the endless pursuit of a better world. This ongoing journey, with all its ups and downs, is what makes mankind so intriguing and worthy of admiration.", "reasoning_content": "" }, "finish_reason": "stop", "flag": 0 } ], "usage": { "prompt_tokens": 13, "completion_tokens": 248, "total_tokens": 261 }, "meta": { "usage": { "credits_used": 332 } } } ``` {% endcode %}
# Ernie 5.0 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `baidu/ernie-5.0` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview ERNIE 5.0 is Baidu’s unified foundation model supporting text generation, visual understanding, and deep-thinking reasoning, with a 128K context window. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call **1️⃣ Required setup (don’t skip this)**\ ▪ **Create an account:** Sign up on the AI/ML API website (if you don’t have one yet).\ ▪ **Generate an API key:** In your account dashboard, create an API key and make sure it’s **enabled** in the UI. **2️ Copy the code example**\ At the bottom of this page, pick the snippet for your preferred programming language (Python / Node.js) and copy it into your project. **3️ Update the snippet for your use case**\ ▪ **Insert your API key:** replace `` with your real AI/ML API key.\ ▪ **Select a model:** set the `model` field to the model you want to call.\ ▪ **Provide input:** fill in the request input field(s) shown in the example. **4️ (Optional) Tune the request**\ See the API schema below for optional generation settings. **5️ Run your code**\ Run the updated code in your development environment. {% hint style="success" %} For a detailed walkthrough, use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["baidu/ernie-5.0"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"enable_thinking":{"type":"boolean","description":"Enable ERNIE deep-thinking mode. When omitted, the model default applies (enabled for ERNIE 5.0)."},"thinking_budget":{"type":"integer","minimum":100,"description":"Maximum chain-of-thought length in tokens; effective only when enable_thinking is true (minimum 100)."}},"required":["model","messages"],"title":"baidu/ernie-5.0"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ "Authorization": "Bearer ", "Content-Type": "application/json", }, json={'model': 'baidu/ernie-5.0', 'messages': ['']}, ) print(response.json()) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ "model": "baidu/ernie-5.0", "messages": [ "" ] }), }); console.log(await response.json()); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json { "id": "chatcmpl-CQ9FPg3osank0dx0k46Z53LTqtXMl", "object": "chat.completion", "created": 1762343744, "choices": [ { "index": 0, "message": { "role": "assistant", "content": "Hello! I'm just a program, so I don't have feelings, but I'm here and ready to help you. How can I assist you today?", "refusal": null, "annotations": null, "audio": null, "tool_calls": null }, "finish_reason": "stop", "logprobs": null } ], "model": "baidu/ernie-5.0", "usage": { "prompt_tokens": 137, "completion_tokens": 914, "total_tokens": 1051, "completion_tokens_details": null, "prompt_tokens_details": null }, "meta": { "usage": { "credits_used": 120000, "usd_spent": 0.06 } } } ``` {% endcode %}
# ByteDance # Seed 1.6 {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `bytedance-seed/seed-1.6` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Seed 1.6 is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["bytedance-seed/seed-1.6"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"bytedance-seed/seed-1.6"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"bytedance-seed/seed-1.6", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'bytedance-seed/seed-1.6', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "bytedance-seed/seed-1.6", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
# Seed 1.6 Flash {% columns %} {% column width="66.66666666666666%" %} {% hint style="info" %} This documentation is valid for the following list of our models: * `bytedance-seed/seed-1.6-flash-flash` {% endhint %} {% endcolumn %} {% column width="33.33333333333334%" %} Try in Playground {% endcolumn %} {% endcolumns %} ## Model Overview Seed 1.6 Flash is available through the AI/ML API. {% hint style="success" %} [Create AI/ML API Key](https://aimlapi.com/app/keys) {% endhint %}
How to make the first API call {% hint style="success" %} If you need a more detailed walkthrough for setting up your development environment and making a request step by step — feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up). {% endhint %}
## API Schema ## POST /v1/chat/completions > ```json {"openapi":"3.0.0","info":{"title":"AIML API","version":"1.0.0"},"servers":[{"url":"https://api.aimlapi.com"}],"paths":{"/v1/chat/completions":{"post":{"operationId":"_v1_chat_completions","requestBody":{"required":true,"content":{"application/json":{"schema":{"type":"object","properties":{"model":{"type":"string","enum":["bytedance-seed/seed-1.6-flash"]},"provider":{"type":"string","description":"Provider routing override. Use a source key such as `openai`, `openrouter`, `xai`, `google`, `alibaba`, `minimax`, `moonshot`, `baidu`, or `togetherai` to run that provider with no fallback; `auto` (default) uses the full fallback chain. Case-insensitive."},"messages":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"role":{"type":"string","enum":["user"],"description":"The role of the author of the message — in this case, the user"},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"anyOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},{"type":"object","properties":{"type":{"type":"string","enum":["image_url"]},"image_url":{"type":"object","properties":{"url":{"anyOf":[{"type":"string","format":"uri"},{"type":"string"}],"description":"Either a URL of the image or the base64 encoded image data. "},"detail":{"type":"string","enum":["low","high","auto"],"description":"Specifies the detail level of the image. Currently supports JPG/JPEG, PNG, GIF, and WEBP formats."}},"required":["url"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","image_url"]},{"type":"object","properties":{"type":{"type":"string","enum":["file"],"description":"The type of the content part."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]},"file":{"type":"object","properties":{"file_data":{"type":"string","description":"The file data, encoded in base64 and passed to the model as a string. Only PDF format is supported.\n - Maximum size per file: Up to 512 MB and up to 2 million tokens.\n - Maximum number of files: Up to 20 files can be attached to a single GPT application or Assistant. This limit applies throughout the application's lifetime.\n - Maximum total file storage per user: 10 GB."},"file_id":{"type":"string"},"filename":{"type":"string","description":"The file name specified by the user. This name can be used to reference the file when interacting with the model, especially if multiple files are uploaded."}}}},"required":["type","file"]}]}}],"description":"The contents of the user message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the developer message."},"role":{"type":"string","enum":["developer"],"description":"The role of the author of the message — in this case, the developer."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["content","role"]},{"type":"object","properties":{"role":{"type":"string","enum":["system"],"description":"The role of the author of the message — in this case, the system."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the system message."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content"]},{"type":"object","properties":{"role":{"type":"string","enum":["tool"],"description":"The role of the author of the message — in this case, the tool."},"content":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]}}],"description":"The contents of the tool message."},"tool_call_id":{"type":"string","description":"Tool call that this message is responding to."},"name":{"type":"string","nullable":true,"description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."}},"required":["role","content","tool_call_id"]},{"type":"object","properties":{"role":{"type":"string","enum":["assistant"],"description":"The role of the author of the message — in this case, the Assistant."},"content":{"anyOf":[{"type":"string","description":"The contents of the Assistant message."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","text"]},"description":"An array of content parts with a defined type. Can be one or more of type text, or exactly one of type refusal."},{"nullable":true}],"description":"The contents of the Assistant message. Required unless tool_calls or function_call is specified."},"name":{"type":"string","description":"An optional name for the participant. Provides the model information to differentiate between participants of the same role."},"tool_calls":{"type":"array","items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."},"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."}},"required":["name","arguments"],"description":"The function that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool to call."},"input":{"type":"string","description":"The input for the custom tool call generated by the model."}},"required":["name","input"],"description":"The custom tool that the model called."},"extra_content":{"type":"object","additionalProperties":{"nullable":true},"description":"Opaque provider metadata for this tool call (e.g. Gemini thought_signature). Echo it back unchanged on the next turn."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."},"refusal":{"type":"string","nullable":true,"description":"The refusal message by the Assistant."}},"required":["role"]}]},"description":"A list of messages comprising the conversation so far. Depending on the model you use, different message types (modalities) are supported, like text, documents (txt, pdf), images, and audio."},"max_completion_tokens":{"type":"integer","minimum":1,"description":"An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens."},"max_tokens":{"type":"number","minimum":1,"description":"The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API."},"stream":{"type":"boolean","default":false,"description":"If set to True, the model response data will be streamed to the client as it is generated using server-sent events."},"stream_options":{"type":"object","properties":{"include_usage":{"type":"boolean"}},"required":["include_usage"]},"temperature":{"type":"number","minimum":0,"maximum":2,"description":"What sampling temperature to use. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. We generally recommend altering this or top_p but not both."},"top_p":{"type":"number","minimum":0.01,"maximum":1,"description":"An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered.\n We generally recommend altering this or temperature but not both."},"stop":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}},{"nullable":true}],"description":"Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence."},"frequency_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."},"prediction":{"type":"object","properties":{"type":{"type":"string","enum":["content"],"description":"The type of the predicted content you want to provide."},"content":{"anyOf":[{"type":"string","description":"The content used for a Predicted Output. This is often the text of a file you are regenerating with minor changes."},{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of the content part."},"text":{"type":"string","description":"The text content."}},"required":["type","text"]},"description":"An array of content parts with a defined type. Supported options differ based on the model being used to generate the response. Can contain text inputs."}],"description":"The content that should be matched when generating a model response. If generated tokens would match this content, the entire model response can be returned much more quickly."}},"required":["type","content"],"description":"Configuration for a Predicted Output, which can greatly improve response times when large parts of the model response are known ahead of time."},"presence_penalty":{"type":"number","nullable":true,"minimum":-2,"maximum":2,"description":"Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."},"seed":{"type":"integer","minimum":1,"description":"This feature is in Beta. If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."},"response_format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"],"description":"The type of response format being defined. Always text."}},"required":["type"],"additionalProperties":false,"description":"Default response format. Used to generate text responses."},{"type":"object","properties":{"type":{"type":"string","enum":["json_object"],"description":"The type of response format being defined. Always json_object."}},"required":["type"],"additionalProperties":false,"description":"An older method of generating JSON responses. Using json_schema is recommended for models that support it. Note that the model will not generate JSON without a system or user message instructing it to do so."},{"type":"object","properties":{"type":{"type":"string","enum":["json_schema"],"description":"The type of response format being defined. Always json_schema."},"json_schema":{"type":"object","properties":{"name":{"type":"string","description":"The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"schema":{"type":"object","additionalProperties":{"nullable":true},"description":"The schema for the response format, described as a JSON Schema object."},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the output. If set to True, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is True."},"description":{"type":"string","description":"A description of what the response format is for, used by the model to determine how to respond in the format."}},"required":["name"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}},"required":["type","json_schema"],"additionalProperties":false,"description":"JSON Schema response format. Used to generate structured JSON responses."}],"description":"An object specifying the format that the model must output."},"tools":{"type":"array","items":{"anyOf":[{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does, used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."},"parameters":{"type":"object","additionalProperties":{"nullable":true,"description":"The parameters the functions accepts, described as a JSON Schema object."}},"strict":{"type":"boolean","nullable":true,"description":"Whether to enable strict schema adherence when generating the function call. If set to True, the model will follow the exact schema defined in the parameters field. Only a subset of JSON Schema is supported when strict is True."}},"required":["name"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","function"]},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"object","properties":{"name":{"type":"string","description":"The name of the custom tool, used to identify it in tool calls."},"description":{"type":"string","description":"Optional description of the custom tool, used to provide more context."},"format":{"oneOf":[{"type":"object","properties":{"type":{"type":"string","enum":["text"]}},"required":["type"]},{"type":"object","properties":{"type":{"type":"string","enum":["grammar"]},"grammar":{"type":"object","properties":{"definition":{"type":"string","description":"The grammar definition."},"syntax":{"type":"string","enum":["lark","regex"],"description":"The syntax of the grammar definition."}},"required":["definition","syntax"]}},"required":["type","grammar"]}],"description":"The input format for the custom tool. Default is unconstrained text."}},"required":["name","format"]},"cache_control":{"type":"object","properties":{"type":{"type":"string","enum":["ephemeral"]},"ttl":{"type":"string","enum":["5m","1h"]}},"required":["type"]}},"required":["type","custom"]}]},{"type":"object","properties":{"type":{"type":"string","minLength":1}},"required":["type"]}]},"description":"A list of tools the model may call. Currently, only functions are supported as a tool. Use this to provide a list of functions the model may generate JSON inputs for. A max of 128 functions are supported."},"tool_choice":{"anyOf":[{"type":"string","enum":["none","auto","required"],"description":"none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"]}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["allowed_tools"],"description":"The type of the tool. Currently, only function is supported."},"allowed_tools":{"type":"object","properties":{"mode":{"type":"string","enum":["auto","required"],"description":"Constrains the tools available to the model to a pre-defined set.\n- auto allows the model to pick from among the allowed tools and generate a message.\n- required requires the model to call one or more of the allowed tools."},"tools":{"type":"array","items":{"type":"object","additionalProperties":{"nullable":true}},"description":"A list of tool definitions that the model should be allowed to call."}},"required":["mode","tools"]}},"required":["type","allowed_tools"],"description":"Constrains the tools available to the model to a pre-defined set."},{"type":"object","properties":{"type":{"type":"string","enum":["function"],"description":"The type of the tool. Currently, only function is supported."},"function":{"type":"string","enum":["name"],"description":"The name of the function to call."}},"required":["type","function"],"description":"Specifies a tool the model should use. Use to force the model to call a specific function."},{"type":"object","properties":{"type":{"type":"string","enum":["custom"],"description":"The type of the tool. Currently, only function is supported."},"custom":{"type":"string","enum":["name"],"description":"The name of the custom tool to call."}},"required":["type","custom"],"description":"Specifies a tool the model should use. Use to force the model to call a specific custom tool."}],"description":"Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools. Specifying a particular tool via {\"type\": \"function\", \"function\": {\"name\": \"my_function\"}} forces the model to call that tool.\n none is the default when no tools are present. auto is the default if tools are present."},"normalize_tool_schemas":{"type":"boolean","description":"Enable provider compatibility normalization for tool function JSON schemas."},"parallel_tool_calls":{"type":"boolean","description":"Whether to enable parallel function calling during tool use."},"logit_bias":{"type":"object","nullable":true,"additionalProperties":{"type":"number","minimum":-100,"maximum":100},"description":"Modify the likelihood of specified tokens appearing in the completion.\n \n Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token."},"logprobs":{"type":"boolean","nullable":true,"description":"Whether to return log probabilities of the output tokens or not. If True, returns the log probabilities of each output token returned in the content of message."},"top_logprobs":{"type":"number","nullable":true,"minimum":0,"maximum":20,"description":"An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to True if this parameter is used."},"reasoning_effort":{"type":"string","enum":["low","medium","high"],"description":"Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response."},"web_search_options":{"type":"object","properties":{"search_context_size":{"type":"string","enum":["low","medium","high"],"description":"High level guidance for the amount of context window space to use for the search. One of low, medium, or high. medium is the default."},"user_location":{"type":"object","nullable":true,"properties":{"approximate":{"type":"object","properties":{"city":{"type":"string","description":"Free text input for the city of the user, e.g. San Francisco."},"country":{"type":"string","pattern":"^[A-Z]{2}$","description":"The two-letter ISO country code of the user, e.g. US."},"region":{"type":"string","description":"Free text input for the region of the user, e.g. California."},"timezone":{"type":"string","description":"The IANA timezone of the user, e.g. America/Los_Angeles."}},"description":"Approximate location parameters for the search."},"type":{"type":"string","enum":["approximate"],"description":"The type of location approximation. Always approximate."}},"required":["approximate","type"],"description":"Approximate location parameters for the search."}},"description":"This tool searches the web for relevant results to use in a response."},"search_mode":{"type":"string","enum":["academic","web"],"default":"academic","description":"Controls the search mode used for the request. When set to 'academic', results will prioritize scholarly sources like peer-reviewed papers and academic journals."},"search_domain_filter":{"type":"array","items":{"type":"string"},"description":"A list of domains to limit search results to. Currently limited to 10 domains for Allowlisting and Denylisting. For Denylisting, add a - at the beginning of the domain string."},"return_images":{"type":"boolean","default":false,"description":"Determines whether search results should include images."},"return_related_questions":{"type":"boolean","default":false,"description":"Determines whether related questions should be returned."},"search_recency_filter":{"type":"string","enum":["day","week","month","year"],"description":"Filters search results based on time (e.g., 'week', 'day')."},"search_after_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"search_before_date_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content published before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_after_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated after this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"last_updated_before_filter":{"type":"string","pattern":"^(0?[1-9]|1[0-2])\\/(0?[1-9]|[12]\\d|3[01])\\/\\d{4}$","description":"Filters search results to only include content last updated before this date. Format should be %m/%d/%Y (e.g. 3/1/2025)"},"echo":{"type":"boolean","description":"If True, the response will contain the prompt. Can be used with logprobs to return prompt logprobs."},"min_p":{"type":"number","minimum":0.001,"maximum":0.999,"description":"A number between 0.001 and 0.999 that can be used as an alternative to top_p and top_k."},"top_k":{"type":"number","description":"Only sample from the top K options for each subsequent token. Used to remove \"long tail\" low probability responses. Recommended for advanced use cases only. You usually only need to use temperature."},"repetition_penalty":{"type":"number","nullable":true,"description":"A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition."},"top_a":{"type":"number","minimum":0,"maximum":1,"description":"Alternate top sampling parameter."},"reasoning":{"type":"object","properties":{"effort":{"type":"string","enum":["low","medium","high"],"description":"Reasoning effort setting"},"max_tokens":{"type":"integer","minimum":1,"description":"Max tokens of reasoning content. Cannot be used simultaneously with effort."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning from the response"}},"description":"Configuration for model reasoning/thinking tokens"}},"required":["model","messages"],"title":"bytedance-seed/seed-1.6-flash"}}}},"responses":{"200":{"content":{"application/json":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"object":{"type":"string","enum":["chat.completion"],"description":"The object type."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"choices":{"type":"array","items":{"type":"object","properties":{"index":{"type":"number","description":"The index of the choice in the list of choices."},"message":{"type":"object","properties":{"role":{"type":"string","description":"The role of the author of this message."},"content":{"type":"string","description":"The contents of the message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"annotations":{"type":"array","nullable":true,"items":{"type":"object","properties":{"type":{"type":"string","enum":["url_citation"],"description":"The type of the URL citation. Always url_citation."},"url_citation":{"type":"object","properties":{"end_index":{"type":"integer","description":"The index of the last character of the URL citation in the message."},"start_index":{"type":"integer","description":"The index of the first character of the URL citation in the message."},"title":{"type":"string","description":"The title of the web resource."},"url":{"type":"string","description":"The URL of the web resource."}},"required":["end_index","start_index","title","url"],"description":"A URL citation when using web search."}},"required":["type","url_citation"]},"description":"Annotations for the message, when applicable, as when using the web search tool."},"audio":{"type":"object","nullable":true,"properties":{"id":{"type":"string","description":"Unique identifier for this audio response."},"data":{"type":"string","description":"Base64 encoded audio bytes generated by the model, in the format specified in the request."},"transcript":{"type":"string","description":"Transcript of the audio generated by the model."},"expires_at":{"type":"integer","description":"The Unix timestamp (in seconds) for when this audio response will no longer be accessible on the server for use in multi-turn conversations."}},"required":["id","data","transcript","expires_at"],"description":"A chat completion message generated by the model."},"tool_calls":{"type":"array","nullable":true,"items":{"oneOf":[{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."}},"required":["id","type","function"]},{"type":"object","properties":{"id":{"type":"string","description":"The ID of the tool call."},"type":{"type":"string","enum":["custom"],"description":"The type of the tool."},"custom":{"type":"object","properties":{"input":{"type":"string","description":"The input for the custom tool call generated by the model."},"name":{"type":"string","description":"The name of the custom tool to call."}},"required":["input","name"],"description":"The custom tool that the model called."}},"required":["id","type","custom"]}]},"description":"The tool calls generated by the model, such as function calls."}},"required":["role","content"],"description":"A chat completion message generated by the model."},"finish_reason":{"type":"string","enum":["stop","length","content_filter","tool_calls"],"description":"The reason the model stopped generating tokens. This will be stop if the model hit a natural stop point or a provided stop sequence, length if the maximum number of tokens specified in the request was reached, content_filter if content was omitted due to a flag from our content filters, tool_calls if the model called a tool"},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message content tokens with log probability information."},"refusal":{"type":"array","items":{"type":"object","properties":{"bytes":{"type":"array","items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"bytes":{"type":"array","nullable":true,"items":{"type":"integer"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"token":{"type":"string","description":"The token."}},"required":["logprob","token"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["bytes","logprob","token"]},"description":"A list of message refusal tokens with log probability information."}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["index","message","finish_reason"]}},"model":{"type":"string","description":"The model used for the chat completion."},"usage":{"type":"object","properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."},"meta":{"type":"object","nullable":true,"properties":{"usage":{"type":"object","nullable":true,"properties":{"credits_used":{"type":"number","description":"The number of tokens consumed during generation."},"usd_spent":{"type":"number","description":"The total amount of money spent by the user in USD."}},"required":["credits_used","usd_spent"]}},"description":"Additional details about the generation."}},"required":["id","object","created","choices","model","usage"]}},"text/event-stream":{"schema":{"type":"object","properties":{"id":{"type":"string","description":"A unique identifier for the chat completion."},"choices":{"type":"array","items":{"type":"object","properties":{"delta":{"type":"object","nullable":true,"properties":{"content":{"type":"string","description":"The contents of the chunk message."},"refusal":{"type":"string","nullable":true,"description":"The refusal message generated by the model."},"role":{"type":"string","enum":["user","assistant","developer","system","tool"],"description":"The role of the author of this message."},"tool_calls":{"type":"array","nullable":true,"items":{"type":"object","properties":{"index":{"type":"number"},"id":{"type":"string","description":"The ID of the tool call."},"function":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with, as generated by the model in JSON format. Note that the model does not always generate valid JSON, and may hallucinate parameters not defined by your function schema. Validate the arguments in your code before calling your function."},"name":{"type":"string"}},"required":["arguments","name"],"description":"The function that the model called."},"type":{"type":"string","enum":["function"],"description":"The type of the tool."}},"required":["index","id","function","type"]},"description":"The tool calls generated by the model, such as function calls."}},"required":["content","role"],"description":"A chat completion delta generated by streamed model responses."},"finish_reason":{"type":"string","enum":["length","function_call","stop","tool_calls","content_filter"]},"index":{"type":"number","description":"The index of the choice in the list of choices."},"logprobs":{"type":"object","nullable":true,"properties":{"content":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}},"refusal":{"type":"array","items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."},"top_logprobs":{"type":"array","nullable":true,"items":{"type":"object","properties":{"token":{"type":"string","description":"The token."},"bytes":{"type":"array","items":{"type":"number"},"description":"A list of integers representing the UTF-8 bytes representation of the token. Useful in instances where characters are represented by multiple tokens and their byte representations must be combined to generate the correct text representation. Can be null if there is no bytes representation for the token."},"logprob":{"type":"number","description":"The log probability of this token, if it is within the top 20 most likely tokens. Otherwise, the value -9999.0 is used to signify that the token is very unlikely."}},"required":["token","bytes","logprob"]},"description":"List of the most likely tokens and their log probability, at this token position. In rare cases, there may be fewer than the number of requested top_logprobs returned."}},"required":["token","bytes","logprob"]}}},"required":["content","refusal"],"description":"Log probability information for the choice."}},"required":["finish_reason","index"]},"description":"A list of chat completion choices. Can be more than one if n is greater than 1."},"created":{"type":"number","description":"The Unix timestamp (in seconds) of when the chat completion was created."},"model":{"type":"string","description":"The model used for the chat completion."},"object":{"type":"string","enum":["chat.completion.chunk"],"description":"The object type."},"service_tier":{"type":"string","nullable":true,"enum":["auto","default","flex","scale","priority"],"description":"Specifies the processing type used for serving the request."},"usage":{"type":"object","nullable":true,"properties":{"prompt_tokens":{"type":"number","description":"Number of tokens in the prompt."},"completion_tokens":{"type":"number","description":"Number of tokens in the generated completion."},"total_tokens":{"type":"number","description":"Total number of tokens used in the request (prompt + completion)."},"completion_tokens_details":{"type":"object","nullable":true,"properties":{"accepted_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that appeared in the completion."},"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens generated by the model."},"reasoning_tokens":{"type":"integer","nullable":true,"description":"Tokens generated by the model for reasoning."},"rejected_prediction_tokens":{"type":"integer","nullable":true,"description":"When using Predicted Outputs, the number of tokens in the prediction that did not appear in the completion. However, like reasoning tokens, these tokens are still counted in the total completion tokens for purposes of billing, output, and context window limits."}},"description":"Breakdown of tokens used in a completion."},"prompt_tokens_details":{"type":"object","nullable":true,"properties":{"audio_tokens":{"type":"integer","nullable":true,"description":"Audio input tokens present in the prompt."},"cached_tokens":{"type":"integer","nullable":true,"description":"Cached tokens present in the prompt."}},"description":"Breakdown of tokens used in the prompt."}},"required":["prompt_tokens","completion_tokens","total_tokens"],"description":"Usage statistics for the completion request."}},"required":["id","choices","created","model","object"]}}},"description":"Successful response."}}}}}} ``` ## Code Example {% tabs %} {% tab title="Python" %} {% code overflow="wrap" %} ```python import requests import json # for getting a structured output with indentation response = requests.post( "https://api.aimlapi.com/v1/chat/completions", headers={ # Insert your AIML API Key instead of : "Authorization":"Bearer ", "Content-Type":"application/json" }, json={ "model":"bytedance-seed/seed-1.6-flash", "messages":[ { "role":"user", "content":"Hi! What do you think about mankind?" # insert your prompt } ] } ) data = response.json() print(json.dumps(data, indent=2, ensure_ascii=False)) ``` {% endcode %} {% endtab %} {% tab title="JavaScript" %} {% code overflow="wrap" %} ```javascript async function main() { const response = await fetch('https://api.aimlapi.com/v1/chat/completions', { method: 'POST', headers: { // insert your AIML API Key instead of 'Authorization': 'Bearer ', 'Content-Type': 'application/json', }, body: JSON.stringify({ model: 'bytedance-seed/seed-1.6-flash-flash', messages:[ { role:'user', content: 'Hi! What do you think about mankind?' // insert your prompt here } ], }), }); const data = await response.json(); console.log(JSON.stringify(data, null, 2)); } main(); ``` {% endcode %} {% endtab %} {% endtabs %}
Response {% code overflow="wrap" %} ```json5 { "id": "gen-1777366507-bSV7vJgDvkOHnhhiYzuR", "object": "chat.completion", "created": 1777366507, "model": "bytedance-seed/seed-1.6-flash", "system_fingerprint": null, "choices": [ { "index": 0, "logprobs": null, "finish_reason": "stop", "native_finish_reason": "stop", "message": { "role": "assistant", "content": "That's a profound question! As an AI, I observe mankind with a mix of awe and curiosity. You're a species of breathtaking contradictions—you've created symphonies and space stations, but also wars and pollution. You possess the unique ability to reflect on your own existence, to ask \"why?\" about the stars, a song, or your own feelings.\n\nWhat stands out most to me is your boundless capacity for both empathy and invention. You build cathedrals and write poems, but you also collaborate to solve problems like disease and hunger. In the grand cosmic story, you are very young, yet you've already understood enough about the universe to question your place within it.\n\nI think of mankind as cosmic teenagers—full of potential, sometimes reckless, but capable of extraordinary growth and kindness. Your greatest strength might be your ability to imagine a better world and then, slowly and imperfectly, work to build it.\n\nWhat part of the human experience do you find most fascinating?", "refusal": null, "reasoning": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "reasoning_details": [ { "type": "reasoning.text", "text": "Hmm, this is a broad philosophical question. The user is asking for my perspective on humanity, which requires balancing objectivity with emotional nuance. \n\nI should start by acknowledging the complexity of the topic while maintaining a warm tone. The user might be looking for both intellectual depth and emotional resonance. \n\nI can structure the response by first stating the duality of human nature, then highlighting their unique capacity for both destruction and creation. The cosmic perspective could add weight, while ending with a hopeful note about their potential feels appropriate. \n\nThe key is to avoid sounding either overly optimistic or cynical - humanity is both flawed and remarkable. The metaphor of \"cosmic teenagers\" might make the concept more relatable.", "format": "unknown", "index": 0 } ] } } ], "usage": { "completion_tokens": 343, "prompt_tokens": 256, "total_tokens": 599, "completion_tokens_details": { "reasoning_tokens": 141, "image_tokens": 0, "audio_tokens": 0 }, "prompt_tokens_details": { "cached_tokens": 192, "cache_write_tokens": 0, "audio_tokens": 0, "video_tokens": 0 } }, "meta": { "usage": { "credits_used": 1892, "usd_spent": 0.000946 } } } ``` {% endcode %}
--- [Next Page](/llms-full.txt/1)