Open Source
Other FrameworkAgentsRAGPythonLangGraphChains
LangChain
One of the most widely used frameworks for building LLM-powered agents and chains. Extensive integrations. Raised $125M at $1.25B valuation (Oct 2025). MIT.
Install
$pip install langchain langchain-anthropic About
LangChain is the most widely used framework for building LLM-powered applications. It provides a unified interface for models, a composable chain system for building pipelines, and a rich library of integrations.
LangGraph, built on top of LangChain, adds graph-based state machines for building agentic workflows with loops, branching, and persistent state — the recommended approach for complex multi-step agents.
Getting Started
pip install langchain langchain-anthropic
from langchain_anthropic import ChatAnthropic
from langchain_core.messages import HumanMessage
model = ChatAnthropic(model="claude-sonnet-4-5")
response = model.invoke([HumanMessage(content="Hello!")])
print(response.content)Features
LCEL (LangChain Expression Language) — composable, streaming chain syntax
400+ integrations: models, vector stores, document loaders, tools
LangGraph — graph-based stateful agent framework built on LangChain
RAG toolkit: document loaders, text splitters, embeddings, retrievers
Memory: conversation buffer, summary, vector store-backed
Structured output parsing with Pydantic models
LangSmith integration for tracing, evaluation, and monitoring
Python and JavaScript/TypeScript libraries
Composable chains & agents
RAG & vector store integrations
Memory management
Tool / function calling
Multi-model support
LangSmith observability
References & Links
Project Info
Language Python
License MIT
Stars 102.0k ★
Forks 16.5k
Category Other
Topics
FrameworkAgentsRAGPythonLangGraphChains
Ready to try LangChain?
Free and open source. Install in one command, works with Claude Code, Cursor, Windsurf, and more.
$pip install langchain langchain-anthropic