LangGraph Documentation
by LangChain
The official docs for building stateful, long-running agents with durable execution and human-in-the-loop.
Overview
LangGraph is a low-level orchestration framework and runtime for building, managing, and deploying long-running, stateful agents, maintained by LangChain and used in production by companies such as Klarna, Uber, and J.P. Morgan. The documentation covers its core capabilities: durable execution and checkpointing so agents survive failures and resume where they left off, streaming, human-in-the-loop workflows that let you inspect and modify agent state at any point, and both short-term working memory and long-term cross-session memory. Agents are expressed as graphs (for example via StateGraph), giving explicit control over state, branching, and loops. The docs include Python and JavaScript examples and position LangGraph as the low-level runtime beneath LangChain's higher-level agent abstractions.
At a Glance
- Topic
- Frameworks
- Level
- Intermediate
- Format
- Documentation
- Cost
- Free
- Provider
- LangChain
- Hands-on
- Yes — code/exercises
- Certificate
- None
What You’ll Learn
- ✓How to model agents as stateful graphs with explicit control over state and loops
- ✓How durable execution and checkpointing let agents resume after failures
- ✓How to add human-in-the-loop steps that inspect and modify agent state
- ✓How to give agents short-term and long-term memory
- ✓How to stream intermediate agent output and deploy long-running agents
Highlights
- •Official, continuously updated documentation from the LangGraph maintainers
- •Focuses on production concerns: durability, persistence, and human oversight
- •Low-level control that complements higher-level LangChain abstractions
- •Python and JavaScript support; used in production by major companies
Who It’s For
Best For
- ✓Engineers building production, long-running or stateful agents
- ✓Teams needing human-in-the-loop and durable, resumable agent execution
- ✓Developers who want graph-level control over agent flow
Prerequisites
- •Python (or JavaScript/TypeScript) experience
- •Familiarity with LLM APIs and basic agent concepts
- •Helpful: prior exposure to LangChain
FAQ
What is LangGraph Documentation?
The official documentation for LangGraph, a low-level orchestration framework and runtime for building stateful, long-running AI agents. Aimed at engineers who need fine-grained control over agent state, loops, and human oversight rather than a high-level abstraction.
Is LangGraph Documentation free?
LangGraph Documentation is free to access.
What level is LangGraph Documentation for?
LangGraph Documentation is aimed at a intermediate audience. Recommended background: Python (or JavaScript/TypeScript) experience, Familiarity with LLM APIs and basic agent concepts, Helpful: prior exposure to LangChain.
What will I learn from LangGraph Documentation?
You'll learn: How to model agents as stateful graphs with explicit control over state and loops; How durable execution and checkpointing let agents resume after failures; How to add human-in-the-loop steps that inspect and modify agent state; How to give agents short-term and long-term memory; How to stream intermediate agent output and deploy long-running agents.