DeepSeekHubs

Integrate DeepSeek with LangChain (Python Guide)

Jan 7, 2026

LangChain is the de facto standard framework for building LLM applications. Since DeepSeek V3 is fully compatible with the OpenAI API, we can use the langchain-openai package for seamless integration.

Quick Start (Python)

1. Install Dependencies

pip install -U langchain-openai

2. Standard Integration

We use the ChatOpenAI class, pointing the base_url to DeepSeek.

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model='deepseek-chat', 
    openai_api_key='sk-xxxx', 
    openai_api_base='https://api.deepseek.com',
    max_tokens=1024
)

response = llm.invoke("Hello DeepSeek! Write a Python bubble sort function.")
print(response.content)

For production, store your secrets in a .env file:

OPENAI_API_KEY=sk-xxxx  # Your DeepSeek Key
OPENAI_API_BASE=https://api.deepseek.com

Then simplify your code:

from dotenv import load_dotenv
from langchain_openai import ChatOpenAI

load_dotenv()

# LangChain automatically picks up the env vars
llm = ChatOpenAI(model='deepseek-chat')

Advanced: Streaming

for chunk in llm.stream("Tell me a joke about Python"):
    print(chunk.content, end="", flush=True)

FAQ

Q: AuthenticationError? A: Ensure openai_api_base is exactly https://api.deepseek.com (no /v1 suffix usually, as ChatOpenAI handles it, but try https://api.deepseek.com/v1 if you use an older version).

Q: Function Calling? A: Yes, DeepSeek V3 supports function calling. In LangChain, simply use bind_tools and it works out of the box.

DeepSeekHubs

DeepSeekHubs