Supermemory
Supermemory is a memory and context engine for AI that automatically extracts facts from conversations, builds user profiles, handles knowledge updates and contradictions, and delivers relevant context. It solves the problem of AI assistants forgetting information between conversations. It is for developers building AI products and for individuals who want persistent memory for their AI tools.
✨ Key features
- Extracts facts from conversations and handles temporal changes and contradictions.
- Auto-maintained user profiles with stable facts and recent activity in ~50ms.
- Hybrid search combining RAG and memory in a single query.
- Connectors for Google Drive, Gmail, Notion, OneDrive, GitHub with real-time webhooks.
- Multi-modal extractors for PDFs, images (OCR), videos (transcription), and code.
- Self-hostable with one binary, zero config, and optional offline mode.
🎯 Use cases
- Give AI assistants persistent memory across conversations.
- Build AI agents with user profiles and personalized context.
- Add hybrid search (RAG + memory) to applications.
- Sync external data from Google Drive, Notion, etc. into a searchable knowledge base.
- Run a fully offline memory engine with Ollama.
📦 Installation
🧰 Requirements: Node.js or Python for the API; for self-hosting, a machine with curl or npx; optional API keys for cloud models or connectors.
npm install supermemory # or: pip install supermemory
For self-hosting:
curl -fsSL https://supermemory.ai/install | bashnpx supermemory local🚀 Usage
import Supermemory from "supermemory";
const client = new Supermemory();
// Store a conversation
await client.add({
content: "User loves TypeScript and prefers functional patterns",
containerTag: "user_123",
});
// Get user profile + relevant memories in one call
const { profile, searchResults } = await client.profile({
containerTag: "user_123",
q: "What programming style does the user prefer?",
});
// profile.static → ["Loves TypeScript", "Prefers functional patterns"]
// profile.dynamic → ["Working on API integration"]
// searchResults → Relevant memories ranked by similarity
❓ FAQ
How does Supermemory differ from RAG?
RAG retrieves document chunks statelessly, while Supermemory extracts and tracks facts about users over time, handling updates and contradictions. It runs both together by default.
Can I run Supermemory locally?
Yes, you can self-host with a single binary using curl -fsSL https://supermemory.ai/install | bash or npx supermemory local, and it can run fully offline with Ollama.
What benchmarks does Supermemory top?
It is #1 on LongMemEval, LoCoMo, and ConvoMem, the three major AI memory benchmarks.
What programming languages are supported?
The API is available via npm (TypeScript/JavaScript) and pip (Python).
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.