> For the complete documentation index, see [llms.txt](https://docs.aimlapi.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aimlapi.com/integrations/langchain.md).

# LangChain

Use AI/ML API models in LangChain through the official langchain-aimlapi package.

## About

[LangChain](https://www.langchain.com/) is a framework for building applications and agents on top of language models — chains, tools, retrieval, and memory, with a common interface across model providers.

AI/ML API is an **official LangChain provider**: the integration ships as the `langchain-aimlapi` package on PyPI and is documented in the LangChain docs at [docs.langchain.com](https://docs.langchain.com/oss/python/integrations/providers/aimlapi). One key gives your chains access to chat, text, embedding, image, and video models.

## Quick start

{% stepper %}
{% step %}

## Install the package

```bash
pip install -U langchain-aimlapi
```

{% endstep %}

{% step %}

## Set your API key

```bash
export AIMLAPI_API_KEY=your_key_here
```

{% endstep %}

{% step %}

## Call a model

```python
from langchain_aimlapi import ChatAimlapi

llm = ChatAimlapi(model="openai/gpt-5-5")
print(llm.invoke("Sing a ballad of LangChain.").content)
```

{% endstep %}
{% endstepper %}

## When to use AI/ML API with LangChain

AI/ML API works well with LangChain when you want:

* chat, embedding, image, and video models behind one key and one package
* to swap the model in a chain by changing a string, not a dependency
* an OpenAI-compatible provider that also covers non-OpenAI vendors

## Prerequisites

Before you start, make sure you have:

* Python and LangChain installed
* an AI/ML API key from [aimlapi.com/app/keys](https://aimlapi.com/app/keys)
* a model ID from [aimlapi.com/models](https://aimlapi.com/models)

The base URL is:

```
https://api.aimlapi.com/v1
```

Authentication is read from the `AIMLAPI_API_KEY` environment variable:

```python
import os

os.environ["AIMLAPI_API_KEY"] = "your_key_here"
```

Need a key first? Use [API Key Management](broken://pages/cdd447f8b558d5268491de41a1a02c07771cac9b).

## Available classes

| Class               | Use for          |
| ------------------- | ---------------- |
| `ChatAimlapi`       | chat completions |
| `AimlapiLLM`        | text completions |
| `AimlapiEmbeddings` | embeddings       |
| `AimlapiImageModel` | image generation |
| `AimlapiVideoModel` | video generation |

All classes provide both synchronous and asynchronous APIs.

## Examples

### Chat

```python
from langchain_aimlapi import ChatAimlapi

llm = ChatAimlapi(model="openai/gpt-5-5", temperature=0.7)
llm.invoke("Explain retrieval-augmented generation in three sentences.")
```

### Embeddings

```python
from langchain_aimlapi import AimlapiEmbeddings

embeddings = AimlapiEmbeddings(model="openai/text-embedding-3-small")
embeddings.embed_query("What is the meaning of life?")
```

### Image generation

```python
from langchain_aimlapi import AimlapiImageModel

img = AimlapiImageModel(model="stable-diffusion-v3-medium", size="512x512", n=1)
img.invoke("A serene mountain lake at sunset")
```

### Video generation

```python
from langchain_aimlapi import AimlapiVideoModel

vid = AimlapiVideoModel(model="google/veo-3.0-fast")
vid.invoke("A timelapse of city lights at night")
```

## Model selection

Always pass `model=` explicitly and use the exact catalog ID, including the vendor prefix — `openai/gpt-5-5`, not `gpt-5-5`.

### Good starting models

* `openai/gpt-5-5` — strong general use
* `anthropic/claude-sonnet-5` — coding and long-form work
* `google/gemini-3.6-flash` — lower latency and lower cost
* `openai/text-embedding-3-small` — embeddings

For the full catalog, use [All Model IDs](broken://pages/9db43b2a9ecf64fedb7fcfc447a3ace848669984).

## Config checklist

Make sure these values are set:

* package: `langchain-aimlapi`
* API key variable: `AIMLAPI_API_KEY`
* base URL: `https://api.aimlapi.com/v1` (built in; override with `AIMLAPI_API_BASE`)
* model ID: exact AI/ML API model ID, passed as `model=`

## Troubleshooting

<details>

<summary>I get a model-not-found error</summary>

Pass `model=` explicitly and copy the ID from [aimlapi.com/models](https://aimlapi.com/models). IDs are vendor-prefixed; a bare name such as `gpt-5-5` is not a catalog ID.

</details>

<details>

<summary>I get an auth error</summary>

Check that:

* `AIMLAPI_API_KEY` is set in the environment the Python process actually sees
* the key is valid at [aimlapi.com/app/keys](https://aimlapi.com/app/keys)
* your balance is sufficient at [aimlapi.com/app](https://aimlapi.com/app)

</details>

<details>

<summary>I need a different endpoint</summary>

Set `AIMLAPI_API_BASE` to override the base URL. Leave it unset to use `https://api.aimlapi.com/v1/`.

</details>

## Links

* [All Model IDs](broken://pages/9db43b2a9ecf64fedb7fcfc447a3ace848669984)
* [AI/ML API keys](https://aimlapi.com/app/keys)
* [AI/ML API model catalog](https://aimlapi.com/models)
* [langchain-aimlapi on PyPI](https://pypi.org/project/langchain-aimlapi/)
* [AI/ML API provider page in the LangChain docs](https://docs.langchain.com/oss/python/integrations/providers/aimlapi)
* [langchain-aimlapi repository](https://github.com/aimlapi/langchain-aimlapi)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.aimlapi.com/integrations/langchain.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.