Compose agents and LLM apps from interoperable components
LangChain is an MIT framework for composing models, tools and retrieval into agents and LLM apps. It is not a ChatGPT competitor, and not a synonym for LangGraph or LangSmith.
Updated 2026-09-04. Same-day snapshot of the official repo.
How to Get Started with LangChain
3 steps to install and make one model call.
Install
Official README recommends uv; pip works too.
uv add langchainSet a provider key
Get an API key from a supported provider and set it as in the docs. Model names follow current docs.
export OPENAI_API_KEY=...Follow the official Quickstart
Don't copy old class names from stale blogs. Start at the docs.langchain.com Python Quickstart.
https://docs.langchain.com/oss/python/langchain/quickstartPitfalls
Don't mash three product names into one package
Docs and job posts often say LangChain / LangGraph / LangSmith in one breath. Install and choose from each repo's own docs.
Abstractions age
Framework and integration packages move fast. Pin versions and follow the current Quickstart — don't copy old APIs like `LLMChain` from 2023 blogs.
Canvas PoC vs framework
For a knowledge-Q&A prototype, Dify is often faster. LangChain fits teams who want the flow in code, wired to existing services.
Beside OpenAI Agents SDK & Dify
Based on official docs and licenses. LangChain is a code framework; Dify is an app builder; OpenAI Agents SDK is a thin official library in the OpenAI ecosystem.
| Tool | License | Surface | Positioning | Best for |
|---|---|---|---|---|
| LangChain | MIT | Python / JS framework | Compose agents and LLM apps from components | Code-first, many providers and tools |
| OpenAI Agents SDK | MIT | Official Python SDK | Thin multi-agent, handoffs, tracing | Production agents already on OpenAI |
| Dify | Apache-2.0 + conditions | Visual workspace | RAG / agent apps with less code | SME PoC, product teams |
Official Directory
Official
- GitHub repository
Main Python repo. MIT.
Docs & Install
- Documentation
Install, components and agent tutorials.
- Python Quickstart
Official Python getting started.
- PyPI · langchain
`uv add langchain` or `pip install -U langchain`.
Related
- LangGraph (related)
Controllable agent workflows — not the same package.
- LangChain.js
Official JS/TS counterpart.
Latest Updates
2026-09-04 · GitHub · langchain-ai/langchain
LangChain: interoperable components for agents and LLM appsMIT framework. Official README calls it an agent engineering platform. 140k+ stars. LangGraph / LangSmith are related products, not the same package.
LaoJin's Pick
FAQ
- What is LangChain?
- LangChain is an MIT open-source framework for agents and LLM apps. You compose interoperable models, tools and retrieval. It's a framework, not a model brand.
- How do I install LangChain?
- Official README recommends `uv add langchain`; `pip install -U langchain` also works. Add provider extras from the docs when you need them.
- Is LangChain the same as LangGraph or LangSmith?
- No. LangChain is the component/agent framework; LangGraph is controllable workflows; LangSmith is observability. Official docs treat them as separate products in one ecosystem.
- LangChain vs OpenAI Agents SDK vs Dify?
- Write Python/JS and mix many providers → LangChain. Already on OpenAI, want a thin official SDK → OpenAI Agents SDK. Visual RAG with less code → Dify.
Need to deploy a team-level Agent system?
Open-source frameworks are free, but turning a runtime into an enterprise asset requires access control, RBAC sandboxes, and private RAG integrations. Laojin Global provides commercial-grade architecture and deployment support.
Cite This Hub
We encourage you to capture this hub into your knowledge base, research, or feed it straight to an AI agent under CC BY 4.0.
How this hub stays current
Version and links live in src/content/langchain.ts; star snapshots sync via GitHub API.