qwen3-32b

This documentation is valid for the following list of our models:

• alibaba/qwen3-32b

Try in Playground

Model Overview

A world-class model with comparable quality to DeepSeek R1 while outperforming GPT-4.1 and Claude Sonnet 3.7. Optimized for both complex reasoning and efficient dialogue.

Create AI/ML API Key

How to make the first API call

1️ 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 <YOUR_AIMLAPI_KEY> 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.

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.

API Schema

post

Body

application/json

modelstring · enumRequiredPossible values:

max_completion_tokensinteger · min: 1Optional

An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens.

max_tokensnumber · min: 1Optional

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.

streambooleanOptional

If set to True, the model response data will be streamed to the client as it is generated using server-sent events.

Default: false

tool_choiceany ofOptional

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. none is the default when no tools are present. auto is the default if tools are present.

string · enumOptional

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.

Possible values:

or

parallel_tool_callsbooleanOptional

Whether to enable parallel function calling during tool use.

temperaturenumber · max: 2Optional

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_pnumber · min: 0.01 · max: 1Optional

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. We generally recommend altering this or temperature but not both.

stopany ofOptional

Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.

stringOptional

or

string[]Optional

or

any | nullableOptional

frequency_penaltynumber | nullableOptional

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_penaltynumber | nullableOptional

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.

seedinteger · min: 1Optional

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_formatone ofOptional

An object specifying the format that the model must output.

or

or

logprobsboolean | nullableOptional

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_logprobsnumber | nullableOptional

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_penaltynumber | nullableOptional

A number that controls the diversity of generated text by reducing the likelihood of repeated sequences. Higher values decrease repetition.

enable_thinkingbooleanOptional

Specifies whether to use the thinking mode.

Default: false

thinking_budgetinteger · min: 1Optional

The maximum reasoning length, effective only when enable_thinking is set to true.

Responses

200Success

application/jsontext/event-stream

post

/v1/chat/completions

JavaScriptPythoncURLHTTP

async function main() {
  const response = await fetch('https://api.aimlapi.com/v1/chat/completions', {
    method: 'POST',
    headers: {
      'Authorization': 'Bearer <YOUR_AIMLAPI_KEY>',
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({
      model: 'alibaba/qwen3-32b',
      messages:[
          {
              role:'user',
              content: 'Hello'
          }
      ],
    }),
  });

  const data = await response.json();
  console.log(JSON.stringify(data, null, 2));
}

main();

200Success

{
  "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/qwen3-32b",
  "usage": {
    "prompt_tokens": 137,
    "completion_tokens": 914,
    "total_tokens": 1051,
    "completion_tokens_details": null,
    "prompt_tokens_details": null
  }
}

Code Example #1: Without Thinking and Streaming

enable_thinking must be set to false for non-streaming calls.

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/qwen3-32b",
        "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))

JavaScript

Response

{
  "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
  }
}

Code Example #2: Enable Thinking and Streaming

Python

import requests

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/qwen3-32b",
        "messages":[
            {
                "role":"user",
                "content":"Hello"  # insert your prompt here, instead of Hello
            }
        ],
        "enable_thinking": True, 
        "stream": True
    }
)

print(response.text)

Response

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}}}

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

import requests
import json

response = requests.post(
    "https://api.aimlapi.com/v1/chat/completions",
    headers={
        # Insert your AIML API Key instead of <YOUR_AIMLAPI_KEY>:
        "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())

After running such code, you'll receive only the model's textual output in a clear and structured format:

Response

--- 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? 😊