> ## Documentation Index
> Fetch the complete documentation index at: https://docs.synapsai.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# LangChain integration

> Use SynapsAI Cloud as an OpenAI-compatible backend in LangChain

LangChain's OpenAI wrappers work when you point `OPENAI_API_BASE` to SynapsAI and set the API key. See [Migrate from OpenAI](/guides/migrate-from-openai) for environment variable setup.

## Chat models (recommended)

```bash theme={null}
export OPENAI_API_KEY="$SYNAPSAI_API_KEY"
export OPENAI_API_BASE="https://api.synapsai.cloud/v1"
```

```python theme={null}
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="<YOUR_MODEL_ID>",
    temperature=0.1,
)

response = llm.invoke("Summarize the benefits of short prompt templates.")
print(response.content)
```

## Legacy OpenAI LLM wrapper

Older LangChain versions use `langchain_openai.OpenAI` for completions-style models:

```python theme={null}
from langchain_openai import OpenAI

llm = OpenAI(model="<YOUR_MODEL_ID>", temperature=0.1)
print(llm.invoke("Summarize the benefits of short prompt templates."))
```

<Tip>
  If LangChain throws errors about `api_base`, pass `openai_api_base="https://api.synapsai.cloud/v1"` directly to the model constructor, or ensure environment variables are set before import.
</Tip>

## Embeddings

```python theme={null}
from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings(model="<YOUR_EMBEDDING_MODEL_ID>")
vector = embeddings.embed_query("SynapsAI Cloud private inference")
```

See [Embeddings + FAISS](/guides/integrations-embeddings-faiss) for a retrieval example.