GitHub radar
Mem0 - A Persistent Memory Layer for AI Agents.
An open-source intelligent memory layer for AI agents and assistants. Persist users’ preferences, history of tasks and contexts across all their sessions.
Mem0 is an open-source intelligent memory layer for AI agents and assistants built by YC S24 startup (mem0ai). Mem0 enables agents to remember users’ preferences, history of tasks and contexts across sessions without the need for constant repetition. In April 2026, we introduced a new memory algorithm that can perform ADD-only extractions with a single LLM call (no chains of UPDATE/DELETE), do entity linking between different memories and do multi-signal retrievals based on semantic search, BM25 keyword search and entity search. With our managed platform, LoCoMo benchmark accuracy went up from 71.4 to 92.5 and on LongMemEval went up from 67.8 to 94.4 (with proprietary optimizations on managed platform, SDK performance is comparable, open source). Mem0 requires an LLM to run (by default, we use gpt-5-mini by OpenAI but any LLM API can be used) and is available as pip/npm library, Docker Compose server (self hosted) or as a cloud platform (app.mem0.ai). Mem0 can be integrated with Claude Code, Codex, Cursor, Windsurf using npx skills add.
Why a vibe-coder should care
Say goodbye to repetitive questions from your agent and hello to your assistant that behaves like a co-worker. With the self hosted version, you own all the data yourself. Mem0 is open source under the MIT license.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Deploy Mem0 from here: https://github.com/mem0ai/mem0 — use the Docker Compose option from the README, then install the skill with `npx skills add https://github.com/mem0ai/mem0 --skill mem0` and show me how the agent remembers context between sessions. Ask me for an API key if needed.
Self-hosted option runs via Docker on a regular laptop. Needs an LLM API key (OpenAI by default).
Open on GitHub▌ More finds