Repo Agents Limited info

langchain-ai/langgraph

LangGraph is a low-level orchestration framework for building, managing, and deploying long-running, stateful agents, installed via pip.

  • 43k GitHub stars
  • Python
  • ⚖️ MIT
  • 🎯 Advanced
View on GitHub (opens in a new tab) Setup guide
pip install -U langgraph
langchain-ai/langgraph preview image

What it is

LangGraph is a low-level orchestration framework for building, managing, and deploying long-running, stateful agents. It provides supporting infrastructure for any long-running, stateful workflow or agent, including durable execution, human-in-the-loop, and memory. It can be used standalone or with LangChain products, and is built by LangChain Inc.

Who it's for

  • Developers building long-running, stateful agents or workflows
  • Teams who want low-level control over agent orchestration rather than a higher-level package
  • Developers who want to pair agents with LangSmith for debugging, evals, and deployment

Setup

  1. Install LangGraph

    Install or upgrade the langgraph package with pip.

    bash
    pip install -U langgraph

Pros & cons

Pros

  • Pro:Durable execution lets agents persist through failures and resume from exactly where they left off
  • Pro:Human-in-the-loop support allows inspecting and modifying agent state at any point during execution
  • Pro:Supports both short-term working memory and long-term persistent memory across sessions
  • Pro:Can be used standalone without LangChain, but integrates with LangChain, LangSmith, and Deep Agents

Cons

  • Con:It is low-level; the README points to Deep Agents as a higher-level option for quickly building agents
  • Con:The README contains no code examples or step-by-step quickstart; it defers to external docs
  • Con:Debugging visualization and production deployment features rely on separate LangSmith products

Images

langchain-ai/langgraph preview image
langchain-ai/langgraph preview image · langchain-ai/langgraph on GitHub · MIT
langchain-ai/langgraph preview image