Skip to main content

​
Requirements

LangChain requires Python 3.10 or higher. You can check your Python version: 
python --version
Python versions supported: 3.10, 3.11, 3.12, 3.13, and 3.14

​
Install LangChain

1

Choose your package manager

LangChain can be installed using pip, uv, or poetry:
pip install langchain
2

Verify installation

Verify LangChain is installed correctly:
import langchain
print(langchain.__version__)

​
Core dependencies

When you install langchain, the following core dependencies are automatically included:
  • langchain-core (>=1.2.10) - Base abstractions and interfaces
  • langgraph (>=1.0.8) - Agent orchestration framework
  • pydantic (>=2.7.4) - Data validation and settings management

​
Optional integrations

LangChain follows a modular architecture. Install only the integrations you need:
Install specific model provider packages:
pip install langchain-openai
Install embedding provider packages:
pip install langchain-openai
Install vector store packages:
pip install langchain-chroma
For document processing and chunking:
pip install langchain-text-splitters
For additional community-maintained integrations:
pip install langchain-community
The community package is large and includes many dependencies. Only install if you need integrations not available as standalone packages.

​
Install with extras

You can install LangChain with optional dependencies using extras: 
pip install langchain[openai]

​
Development setup

For development or contributing to LangChain:
1

Clone the repository

git clone https://github.com/langchain-ai/langchain.git
cd langchain
2

Install uv

LangChain uses uv for dependency management:
curl -LsSf https://astral.sh/uv/install.sh | sh
3

Install dependencies

Install all packages in editable mode:
uv sync --all-groups
Or install specific dependency groups:
# Test dependencies only
uv sync --group test

# Lint dependencies
uv sync --group lint
4

Run tests

make test

​
Environment variables

Configure API keys for the integrations you’re using:
.env
# OpenAI
OPENAI_API_KEY=your-openai-key

# Anthropic
ANTHROPIC_API_KEY=your-anthropic-key

# Google GenAI
GOOGLE_API_KEY=your-google-key

# Groq
GROQ_API_KEY=your-groq-key

# LangSmith (optional, for monitoring)
LANGSMITH_API_KEY=your-langsmith-key
LANGSMITH_TRACING=true
Use a .env file with python-dotenv to manage environment variables locally:
from dotenv import load_dotenv
load_dotenv()

​
Verify your setup

Create a simple test script to verify everything works:
test_setup.py
from langchain_core.messages import HumanMessage
from langchain_core.prompts import ChatPromptTemplate

# Test core imports
print("✓ LangChain core imported successfully")

# If you installed a chat model provider:
try:
    from langchain_openai import ChatOpenAI
    model = ChatOpenAI(model="gpt-4o-mini")
    response = model.invoke([HumanMessage(content="Hello!")])
    print(f"✓ Chat model works: {response.content[:50]}")
except ImportError:
    print("! Chat model provider not installed")
except Exception as e:
    print(f"! Error testing chat model: {e}")

print("\n✅ LangChain setup complete!")
Run the test: 
python test_setup.py

​
Next steps

Quickstart

Build your first LLM application

Core concepts

Learn about the framework architecture

Integrations

Explore available providers

Guides

Follow step-by-step guides

​
Troubleshooting

Make sure you installed LangChain in the correct Python environment:
python -m pip install langchain
If you encounter dependency conflicts, try creating a fresh virtual environment:
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install langchain
If you encounter SSL errors when installing:
pip install --trusted-host pypi.org --trusted-host files.pythonhosted.org langchain
Ensure your environment variables are set correctly. Check with:
import os
print(os.getenv("OPENAI_API_KEY"))

Build docs developers (and LLMs) love

Get started for freeTalk to us