Skip to content
WhichAgent

LangChain

Agent framework by LangChain, Inc.

ActiveOpen source · MITVerified Oct 22, 2026
Website Docs Source

Open-source framework with a standard interface over many model providers and a prebuilt agent harness (create_agent) built on LangGraph, available for Python and JavaScript/TypeScript.

Key facts

Type
Agent framework · Library
Pricing
Free
Free tier
The LangChain library is free and MIT-licensed. The optional LangSmith cloud (tracing, evals, deployment) has a free Developer plan with 1 seat and 5k base traces/month.
License
MIT
Models
OpenAI, Anthropic, Google, AWS Bedrock, Azure OpenAI, Mistral, Groq, DeepSeek, xAI, OpenRouter, Local models via Ollama, Many more via partner and community integration packages
Languages
Python, TypeScript
Platforms
macOS, Windows, Linux
Bring your own API key
Yes
Local models
Yes
MCP support
Yes
Self-hostable
Yes
GitHub stars
148,264
Last commit
Oct 22, 2026

Best for

  • Teams that want to swap model providers behind one interface
  • Developers building tool-calling agents quickly from a prebuilt harness
  • Projects that plan to add LangGraph orchestration or LangSmith tracing later

Pros

  • Very large integration ecosystem across model, vector store, and tool providers
  • Official Python and JavaScript/TypeScript libraries
  • Agents run on LangGraph, so they inherit persistence, streaming, and human-in-the-loop support
  • MCP tools supported via langchain-mcp-adapters

Cons

  • Abstraction layers can make low-level debugging harder than calling provider SDKs directly
  • Integrations live in many separate packages with varying maintenance levels
  • History of breaking API changes between major versions

How LangChain compares

LangChain vs LangGraph

They aren't rivals: LangChain's agents are built on top of LangGraph.

Updated Oct 22, 2026

LangChain vs LlamaIndex

For agents, pick LangChain.

Updated Oct 22, 2026

Best LangChain alternatives →

Free, open-source and paid options, sorted by what you need.

Sources (8)