LangGraph

Health v2 · maintenance only

Adoption not part of the score

43532 stars · 7430 forks observed · 2026-10-20

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-28, confidence 0.97

LangGraph is a low‑level Python library for building, managing, and deploying long‑running, stateful agents. It provides durable execution, human‑in‑the‑loop capabilities, and deep debugging support.

Use cases

• Orchestrate complex multi‑step LLM workflows with persistent state
• Build resilient agents that can recover from failures and resume execution
• Add human‑in‑the‑loop checkpoints to LLM‑driven processes
• Debug and trace agent behavior with fine‑grained state inspection
• Create custom agent architectures mixing deterministic and LLM‑driven steps

When to choose

• You need fine‑grained control over agent execution and state management
• Your application requires durable, long‑running agents that survive crashes
• You want to integrate human oversight into LLM workflows
• You need deep observability and debugging of agent behavior

When to avoid

• You prefer a high‑level, out‑of‑the‑box agent solution with minimal coding
• Your project is not Python‑based or requires a non‑Python runtime
• You only need simple stateless API calls to LLMs

Facets

library · maturity active

agent-framework workflow-automation state-management artificial-intelligence large-language-models python cross-platform stateful durable-execution human-in-the-loop langgraph llm-orchestration debugging ai-agents retrieval-augmented-generation

Member repositories

For agents

markdown · JSON · MCP: product_card(name="langchain-ai/langgraph")