Agent frameworks
LangGraph on GitHub: what it does and how to use it with AI agents
LangChain framework for long-running, stateful agents.
Stars, license and last push from the GitHub API, checked 2026-10-26
| Repo | langchain-ai/langgraph |
|---|---|
| Stars | 43,900+ |
| License | MIT |
| Last push | 2026-10-26 |
| Language | Python |
What LangGraph does
LangGraph models an agent as a graph of steps with saved state. That gives you retries, human approval steps and memory across runs. There is a Python and a JavaScript version.
Who it is for
Developers building agents that need to pause, resume or wait for a human, beyond a single chat loop.
How to use LangGraph with Claude Code, Codex or Cursor
Install: pip install -U langgraphStart from a template in the docs.Ask Claude Code or Codex to add a human approval step before any action that sends or spends.Add a checkpointer so runs can resume.
Tip: Have your coding agent draw the graph in plain text first, then code it.
The loadout for this repo
See the full tool stack, prompts and costs in Best AI coding agent setup.
Related repos
- CrewAI: Python framework for teams of role-playing AI agents.
- Browser Use: Open-source Python library for AI agents that use a web browser.
LangGraph FAQ
Is LangGraph free?
The library is open source under MIT. LangChain sells separate hosted products.
LangGraph vs CrewAI?
LangGraph gives you a low-level graph with explicit state. CrewAI organizes agents as a crew with roles and tasks.
Does LangGraph need LangChain?
No. It can be used on its own.
Star counts change daily. Always read a repo's license and code before running it. All repos