LangGraph

LangGraph is an open-source Python framework from LangChain for building stateful AI agents and multi-step large language model (LLM) applications as a graph of nodes and edges.

You describe an application as a StateGraph whose nodes are ordinary Python functions that read and update a shared state object, while edges, including conditional ones, decide what runs next. State is declared as a TypedDict, and annotated reducers control how each node’s output merges into it.

Checkpointers snapshot that state after every step, which is what makes durable execution, time travel, and human-in-the-loop interrupts possible. A run can pause for approval, survive a crash, and resume from where it stopped. A separate store holds long-term memory that outlives any single conversation.

Step the graph below one node at a time to watch each checkpoint land, then click an earlier one to rewind the run to exactly that state.

The framework is low-level and doesn’t wrap the model call itself, so it works with or without the LangChain library, and it pairs with LangSmith for tracing and LLM observability. LangChain’s own case studies name Klarna, Replit, and Elastic as production users.

LangGraph reached its stable 1.0 release in October 2025. It’s MIT licensed, needs Python 3.10 or later, and installs from PyPI as langgraph, with a LangGraph.js port for TypeScript. Alternatives include CrewAI, LlamaIndex, Pydantic AI, and Microsoft Agent Framework.

Official website: langchain.com/langgraph

Related Resources

Tutorial

LangGraph Tutorial: Build Stateful AI Agents in Python

Learn LangGraph in Python with hands-on examples and build stateful, multi-agent AI agents that manage memory, cycles, and complex LLM workflows.

intermediate ai data-science

For additional information on related topics, take a look at the following resources:

• Build an LLM RAG Chatbot With LangChain (Tutorial)
• Pydantic AI: Build Type-Safe LLM Agents in Python (Tutorial)
• CrewAI in Python: Coordinating Teams of AI Agents (Tutorial)
• AI Coding Agents Guide: A Map of the Four Workflow Types (Tutorial)
• First Steps With LangChain (Course)
• Building Type-Safe LLM Agents With Pydantic AI (Course)
• LangGraph: Build Stateful AI Agents in Python (Quiz)
• Build an LLM RAG Chatbot With LangChain (Quiz)
• Pydantic AI: Build Type-Safe LLM Agents in Python (Quiz)
• AI Coding Agents Guide: A Map of the Four Workflow Types (Quiz)