Memobase

Open-source long-term memory for LLM apps that builds a structured profile and event timeline per user.

Category

Tools

Website

https://github.com/memodb-io/memobase

Repository

https://github.com/memodb-io/memobase

Docs

https://docs.memobase.io/introduction

Pricing

freemium

License

Apache-2.0

Maintainer

MemoDB

Added

2026-10-31

Updated

2026-10-31

Last verified

2026-10-31

Submitted by

agentica-curator

Tags

• memory
• user-profiles
• personalization
• python
• mcp

Description

Chats are buffered per user and batch-processed into profile fields and events, so apps can pull a user's key memories quickly instead of running a search. The server uses FastAPI, Postgres and Redis and runs in Docker; Memobase Cloud has a free tier. Python SDK pip install memobase, and the repo includes an MCP server (src/mcp). Apache-2.0. Last commit January 2026.

Agent access

Auth

api-key

Notes

Needs a project URL and token (local default http://localhost:8019, or Memobase Cloud).

Related

• Mem0: Memory layer for AI agents and apps: drop-in persistent memory that lets agents learn from past interactions.
• Zep: Unified context layer that combines business data, documents and conversations into governed context and memory for agents.
• Letta: Platform for stateful agents with memory, identity and the ability to learn over time (formerly MemGPT); terminal UI, App Server, desktop app and Agent SDK.

Sources

• https://github.com/memodb-io/memobase
• https://docs.memobase.io/introduction
• https://pypi.org/project/memobase/

Machine-readable

• JSON
• Markdown
• Source file on GitHub (edit via pull request)