Is Deepseek-R1 API accessed through openai library compatible with Langchain and Langgraph methods?

[leockl]

Looking at Deepseek-R1's API documentation (https://api-docs.deepseek.com/), based on the script below, Deepseek-R1 can been accessed through openai library:

from openai import OpenAI

client = OpenAI(api_key="<DeepSeek API Key>", base_url="https://api.deepseek.com")

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "system", "content": "You are a helpful assistant"},
        {"role": "user", "content": "Hello"},
    ],
    stream=False
)

print(response.choices[0].message.content)

Does that mean Deepseek-R1 API (accessed through openai library) as shown above is compatible with Langchain and Langgraph methods (such as those in from langchain_openai import ChatOpenAI and from langgraph.prebuilt import create_react_agent)?

[dosubot[bot]]

Yes, the Deepseek-R1 API, accessed through the OpenAI library, is compatible with LangChain methods. You can integrate it using the ChatOpenAI class from LangChain by specifying the model and API key. Here's how you can set it up:

from langchain_openai.chat_models import ChatOpenAI

llm = ChatOpenAI(
    model="deepseek-chat",
    api_key="<DeepSeek API Key>",
    openai_api_base="https://api.deepseek.com"
)

This setup allows you to use the Deepseek-R1 API with LangChain's methods, leveraging the shared structure of API calls and message formatting ^[1].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[leockl]

@dosu What is the difference between from langchain_openai.chat_models import ChatOpenAI and from langchain_openai import ChatOpenAI?