Supermemory

Health v2 · maintenance only

Adoption not part of the score

31077 stars · 2734 forks observed · 2026-10-05

What it is AI-extracted, prompt v1, taxonomy v1, 2026-10-05, confidence 0.97

Supermemory is a fast, scalable memory and context engine for AI agents, offered as a hosted API with self‑hosted options. It provides automatic fact extraction, temporal graph storage, hybrid RAG search, and user profile management.

Use cases

• persist user preferences and conversation history across AI chat sessions
• enable RAG over internal documents and real‑time webhooks from Google Drive, Notion, GitHub, etc.
• build AI assistants that can recall project decisions, timelines, and team knowledge
• provide fast (~50 ms) personalized context for large‑language‑model prompts
• integrate multimodal data (PDFs, images, videos) into a searchable knowledge graph

When to choose

• You need a unified memory layer that works for both short‑term chat context and long‑term knowledge bases
• Your AI agents must operate with up‑to‑date user profiles and handle contradictions automatically
• You prefer a solution that can be run locally or self‑hosted for data privacy
• You want built‑in connectors to common productivity tools without writing custom ingestion pipelines

When to avoid

• Your project only requires simple key‑value storage without semantic search
• You need a fully open‑source stack with no external service dependencies
• Latency requirements are stricter than the provided ~50 ms response time
• You are working in environments without Node.js or Python SDK support

Facets

service · maturity active

vector-database search-engine api-framework middleware rag artificial-intelligence large-language-models chatbots cloud self-hosted python memory-engine user-profiles connectors multimodal-extractors mcp company-brain local-run retrieval-augmented-generation data-engineering nodejs web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="supermemoryai/supermemory")