LangGraph
LangChain's MIT-licensed framework for building agents as graphs, with checkpoint-based persistence.
0 members list LangGraph, unchanged over the last 12 weeks, data as of 17 November 2026
Members who list LangGraph in their Stack, by week.
Members who list LangGraph
Be the first to add LangGraph
Nobody on Stackness runs LangGraph yet. Add it to your stack and you're the first member running it.
About LangGraph
LangGraph is a framework for building agents as computational graphs, developed by langchain-ai and licensed under MIT. It provides developers with a low-level orchestration runtime for constructing, managing, and deploying long-running stateful agents. The framework enables fine-grained control by combining deterministic, hand-coded logic with steps driven by language models within the same graph structure. This allows developers to create custom agents tailored to specific application requirements.
LangGraph focuses on core capabilities essential for agent orchestration, including durable execution that persists state through checkpoints, streaming support, and human-in-the-loop interaction patterns. The framework is built in Python and can be used independently or alongside LangChain components for integrating models and tools. It is designed for developers who need low-level control over agent behavior and prefer building bespoke solutions rather than using prebuilt agent architectures.
Checked 11 November 2026.
Nobody on Stackness runs LangGraph yet.
Sign up and add itPosts about LangGraph
- Agent memory format standard: none exists, so keep what must port in the repo
No agent memory format standard exists as of November 2026. Keep what must survive a switch in AGENTS.md and skills, and treat tool memory as a cache.
- What is a durable agent harness? Pi Durable, Pi 1.0 and the slot it takes in an agent stack
A durable harness commits every step of an agent run to storage, so a crash resumes mid-turn. What Pi Durable shipped on 1 October, and what is only claimed.