# Supermemory

A memory and context engine for AI applications that extracts facts from conversations, maintains user profiles, and combines retrieval-augmented generation with memory in a single API.

Supermemory is a memory and context layer for AI applications. It automatically extracts facts from conversations, builds and maintains user profiles, handles updates and contradictions in what it has learned, and forgets information that has expired, then supplies the right context back to an application on request.

It combines full retrieval-augmented generation with memory in a single hybrid search call, includes connectors for services such as Google Drive, Gmail, Notion, OneDrive, and GitHub, and can extract from PDFs, images, videos, and code. It is available as a hosted API, as plugins and an MCP server that give coding agents persistent memory across sessions, or as a self-hosted binary that can run fully offline.

It is written in TypeScript and released under the MIT license.

Website: https://supermemory.ai/docs