LangGraph
LangChainLow-level orchestration framework from LangChain for building stateful, multi-actor agent applications. Unlike higher-level frameworks it exposes an explicit graph/state-machine model (nodes, edges, persistence, human-in-the-loop checkpoints), the production choice when you need fine-grained control over loops, branching, and durable resumable execution.
MIT core; commercial LangGraph Platform / LangSmith is the managed layer (open-core). ~35k stars, ~58M PyPI downloads/month (heavily transitive via LangChain).
Openness
4 high confidence- license
- MIT(core)
- commercial
- LangGraph-Platform/LangSmith
- source
- public
MIT-licensed self-hostable core with a commercial managed deployment platform.
- https://github.com/langchain-ai/langgraph recorded 2026-06-17
MIT, ~35k stars, low-level agent orchestration
Adoption
5 high confidence~58M PyPI downloads/month (directional - largely transitive via LangChain).
- https://pypistats.org/packages/langgraph recorded 2026-06-17
~58.6M downloads/month
Capability
5 high confidenceDeepest durable-execution control among the orchestration peers.
- https://github.com/langchain-ai/langgraph recorded 2026-06-17
feature set: graph control flow, persistence, HITL
Unchanged since 2026-06-17 (last edited, not re-checked)