Open SourceFramework13k
LangGraph
Build stateful, multi-actor agent applications as controllable graphs.
LangGraph is a low-level orchestration framework from the LangChain team for building stateful, multi-agent applications modeled as graphs of nodes and edges. It adds durable execution, persistent memory, human-in-the-loop checkpoints, and cyclic control flow that plain chains cannot express. It is widely used to build production agent runtimes where reliability, branching, and state management matter.
Repository
langchain-ai/langgraphLanguage
PythonWhat you'd build with it
- Modeling a multi-step agent workflow as an explicit, debuggable state graph
- Adding human-in-the-loop approval and checkpointed resume to long-running agents
- Coordinating multiple specialized agents that hand off to each other
Tags
agentsorchestrationgraphstate