Skip to content

LangGraph

Back to Overview: Agents

What Is LangGraph

LangGraph extends LangChain with a graph-based execution model — agents are defined as nodes and edges in a directed graph, enabling cyclical workflows, persistent state, and fine-grained control over agent behavior.

When to Use LangGraph

  • Complex agents that need to loop, branch, or retry steps
  • Long-horizon tasks requiring persistent state across steps
  • Multi-agent systems where you need explicit control flow
  • When you need checkpointing and human-in-the-loop interruptions

Chapter Map

File Topic
01 — LangGraph Fundamentals Graph model, nodes, edges, state
02 — Simple Agent Basic graph agent
03 — Complex Agent Stateful, multi-step graph agent

Back to Agents Index