LangGraph Mode | OpenMemory

Advanced guide on using OpenMemory as a state persistence layer for LangGraph agents.

LangGraph Mode

OpenMemory can operate in a special "LangGraph Mode" where it acts as the persistence layer for your LangGraph agents.

Why use OpenMemory with LangGraph?

Standard LangGraph checkpointers (like SQLiteSaver) just save the state blob. OpenMemory does more:

  1. Semantic Indexing: It indexes the content of the state, allowing you to search for past states by meaning.
  2. Temporal Graph: It maps state transitions to a timeline.

Configuration

To enable this mode, use the OpenMemorySaver class provided in the Python SDK.

from langgraph.checkpoint.openmemory import OpenMemorySaver

See the LangGraph API Reference for implementation details.

© 2025 OpenMemory · MIT License