supermemoryai/supermemory

Memory and context engine + app that is extremely fast, scalable, and can be run fully locally. The Memory API for the AI era.

What it solves

AI assistants typically forget everything between conversations. Supermemory provides a persistent memory and context layer that allows AI to remember user preferences, past discussions, and project details over time, preventing the need to repeat information in every new session.

How it works

Supermemory acts as a context engine that sits between the AI tool and the LLM. It automatically extracts facts from conversations, builds and maintains user profiles (splitting them into stable facts and recent activity), and handles knowledge updates or contradictions (e.g., updating a city of residence). It combines traditional RAG (retrieving document chunks) with personalized memory (tracking user facts) in a single hybrid search query. It also includes connectors to sync data from external sources like Google Drive, Notion, and GitHub, and multi-modal extractors for PDFs, images, and videos.

Who it’s for

  • AI Tool Users: Individuals who want a "personal brain" for their AI assistants via a browser extension, app, or MCP server.
  • Developers: Those building AI agents or apps who need a single API to handle memory, RAG, user profiles, and data connectors without managing vector databases or embedding pipelines.
  • Self-hosters: Users who want to run the memory engine locally and offline using tools like Ollama.

Highlights

  • High Performance: Ranked #1 on major AI memory benchmarks (LongMemEval, LoCoMo, ConvoMem).
  • User Profiles: Maintains auto-updated profiles with ~50ms retrieval time for instant context injection.
  • Hybrid Search: Merges knowledge base documents and personalized user context in one query.
  • Broad Integration: Supports MCP (Model Context Protocol), various AI frameworks (Vercel AI SDK, LangChain, etc.), and multiple IDEs (Cursor, VS Code).
  • Local Mode: Offers a zero-config local binary for fully offline operation.

Related

  • Project
    MemoriLabs/Memori

    A persistent, structured memory layer for AI agents and LLMs that automatically captures and recalls user preferences and agent actions to reduce token costs and maintain context across sessions.

  • Project
    beilusaiying/always-accompany

    A multi-purpose AI agent framework focused on context and attention mechanisms, providing layered memory and dynamic retrieval for long-term companionship, coding, and work.

  • Project
    MemMachine/MemMachine

    An open-source long-term memory layer for AI agents that enables persistent storage and recall of user preferences and conversational context across sessions.

  • Project
    buildingjoshbetter/TrueMemory

    A local-first long-term memory system for AI agents that automatically captures and updates user preferences and facts across sessions to eliminate AI amnesia.

  • Project
    supermemoryai/opencode-supermemory

    A plugin for OpenCode that adds persistent, long-term memory to AI agents, allowing them to remember project details and user preferences across sessions.