AI Agent Platforms / All platforms / LangGraph
LangGraph
Low-level orchestration framework for long-running, stateful agents with explicit graph control.
LangGraph is a low-level framework for building, managing, and deploying long-running, stateful agents. Developers use it when they need explicit control over agent execution, persistence, and human oversight in production systems.
| Category | Agent builders and frameworks |
| License | MIT |
| Hosting | library for self-managed applications; vendor platform available |
| Repository | github.com/langchain-ai/langgraph |
| Website | docs.langchain.com/oss/python/langgraph |
| Last verified | 2026-09-02 |
Documented features
- Persists agent state so runs survive failures and resume where they left off
- Supports long-running agent executions
- Allows inspecting and modifying agent state at any point for human-in-the-loop oversight
- Provides short-term working memory for ongoing reasoning
- Provides long-term persistent memory across sessions
- Integrates with LangSmith for tracing execution paths and state transitions
- Offers deployment infrastructure for stateful, long-running workflows via the vendor platform
- Installs as a Python library via pip
Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.