Skip to main content
This integration connects DeepSeek’s models to LangChain.

​
Installation

pip install -U langchain-deepseek

​
Setup

Set your DeepSeek API key as an environment variable: 
export DEEPSEEK_API_KEY="your-api-key"
Get your API key from platform.deepseek.com.

​
Usage

from langchain_deepseek import ChatDeepSeek

model = ChatDeepSeek(
    model="deepseek-chat",
    temperature=0,
    max_tokens=None,
)

messages = [
    ("system", "You are a helpful assistant."),
    ("human", "What is the capital of France?"),
]

response = model.invoke(messages)
print(response.content)

​
Streaming

for chunk in model.stream(messages):
    print(chunk.content, end="")

​
API Reference

​
ChatDeepSeek

ChatDeepSeek extends ChatOpenAI and inherits all of its parameters. It’s preconfigured to use DeepSeek’s API endpoint.
​
model
str
required
Name of DeepSeek model to use (e.g., deepseek-chat, deepseek-reasoner).
​
temperature
float
default:"1"
Sampling temperature. Controls randomness in generation.
​
max_tokens
int | None
default:"None"
Maximum number of tokens to generate.
​
timeout
float | None
default:"None"
Timeout for requests in seconds.
​
max_retries
int
default:"2"
Maximum number of retries for failed requests.
​
api_key
str | None
default:"None"
DeepSeek API key. If not provided, reads from DEEPSEEK_API_KEY environment variable.

​
Supported Models

  • DeepSeek-V3: Latest flagship model with excellent performance
  • DeepSeek-Chat: General-purpose chat model
  • DeepSeek-Reasoner: Specialized for complex reasoning tasks
See platform.deepseek.com/api-docs for the latest models.

​
Features

  • Text generation
  • Function/tool calling
  • Reasoning mode (DeepSeek-Reasoner)
  • Streaming
  • Async support
  • Competitive pricing
  • Support for long context windows
DeepSeek models are known for their strong reasoning capabilities and cost-effectiveness. The DeepSeek-Reasoner model includes a thinking process similar to OpenAI’s o1 series.

​
Reasoning Example

For models with reasoning capabilities: 
model = ChatDeepSeek(model="deepseek-reasoner")

response = model.invoke([
    ("human", "Solve this step by step: If a train leaves at 2pm going 60mph, and another leaves at 3pm going 80mph, when will they meet?")
])

# Access reasoning content if available
if hasattr(response, 'additional_kwargs'):
    reasoning = response.additional_kwargs.get('reasoning_content')
    if reasoning:
        print("Reasoning:", reasoning)

print("Answer:", response.content)

Build docs developers (and LLMs) love

Get started for freeTalk to us