LangMem
MemoryMIT
LangMem is a Python library from LangChain that lets an agent record and search memories during a conversation and consolidate them in the background. It offers functional primitives that work with any storage system and a native integration with LangGraph's storage layer.
The README describes memory management tools that an agent calls in the hot path, a background memory manager that extracts and updates knowledge, and native use of the LangGraph long-term memory store.

Pinout
- MCP server: not checked yet
- HTTP API: not checked yet
- command line: not checked yet
- Python SDK: supported, verified
- TypeScript SDK: not checked yet
- OpenAPI description: not checked yet
| Interface | Status | Source we read | Read on |
|---|---|---|---|
| MCP server | Not checked yet | — | — |
| HTTP API | Not checked yet | — | — |
| Command line | Not checked yet | — | — |
| Python SDK | Supported, verified | README, Installation (pip install -U langmem) | |
| TypeScript SDK | Not checked yet | — | — |
| OpenAPI description | Not checked yet | — | — |
Facts
| License | MIT. GitHub API, repository license field, read . |
|---|---|
| Main language | Python (as detected by GitHub). GitHub API, repository record, read . |
| Publisher | langchain-ai (GitHub organization). GitHub API, repository record, read . |
| Last push to the repository | 2026-11-13. GitHub API, repository record, read . |
| LangGraph integration | Native integration with the LangGraph long-term memory store. README in the repository, read . |
| Last commit | . GitHub API, latest commit on the default branch of langchain-ai/langmem, read . |
Call example
pip install -U langmemLimits
- The README setup asks for an API key for an LLM provider.
Strengths
- Its memory tools plug directly into a LangGraph agent created with create_react_agent, as shown in the README.
Alternatives to LangMem
All memory tools for agents
Agent memory system with retain, recall and reflect operations.

Open-source memory layer storing agent context across sessions and models.

Memory layer that wraps LLM clients to store and recall agent state.

Stateful agents whose memory and identity persist across sessions and machines.
Last check: . Method: public repository and docs read automatically, page and day written down. Next check due within 90 days. How we check
Is this your tool? Write to hello@agent-tools.directory from an address on its domain to correct a pin or a fact.