# CaviraOSS/OpenMemory

Local persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc.

Repository: https://github.com/CaviraOSS/OpenMemory
Canonical: https://ross.abutalabs.com/products/openmemory
Homepage: https://openmemory.cavira.app
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: ai, ai-agents, ai-infrastructure, ai-memory, artificial-intelligence, cognitive-architecture, embeddings, gemini, llm, long-term-memory, memory, memory-engine, memory-retrieval, ollama, openai, openmemory, rag, supermemory, vector-database, one-line
Last push: 2026-08-25T17:10:18+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 60, longevity 22
- inputs: {"age_days": 318, "days_push": 8, "days_rel": 264, "gap_med": 6.0, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4466, forks 501 (observed 2026-08-28T04:08:50.876840+00:00)

## What it is
OpenMemory is a local-first, self-hosted cognitive memory engine that gives LLM applications and AI agents persistent long-term memory, with multi-sector embeddings, temporal knowledge graphs, and adaptive memory decay. It ships as Python and Node SDKs (standalone SQLite mode or backend server mode) and integrates with OpenAI, LangChain, CrewAI, AutoGen, and MCP clients like Claude Desktop and Cursor.

## Use cases
- give my AI agent persistent memory across sessions
- remember user preferences in a chatbot between conversations
- add long-term memory to Claude Desktop via MCP
- self-hosted alternative to Mem0 or Supermemory
- store and recall episodic and semantic memories for LLM apps
- give each Streamlit app user persistent memory by user_id
- ingest PDFs and web pages into an agent's memory
- reduce token costs by not re-sending context every session

## When to choose
- you want local-first, self-hosted memory with full data ownership and no vendor API calls
- you need explainable recall traces and multi-sector (episodic/semantic/emotional) memory rather than plain vector search
- you use LangChain, CrewAI, AutoGen, or MCP clients and want drop-in memory integration
- you want per-user memory scoping in multi-user applications

## When to avoid
- you need a battle-tested production system - the project is being rewritten with expected breaking changes and bugs
- you just need a plain vector database or simple RAG over documents
- you require guaranteed stability, long-term support, or enterprise SLAs
- you want a fully managed hosted memory service

## Facets
- artifact type: library
- maturity: experimental
- function: vector-database, rag, llm-inference, agent-framework, mcp, database, sdk
- domain: artificial-intelligence, large-language-models, developer-tools, self-hosted
- platform: python, cross-platform, self-hosted
- tags: long-term-memory, memory-engine, cognitive-architecture, local-first, sqlite, postgres, temporal-knowledge-graph, multi-sector-embeddings, memory-decay, langchain, crewai, autogen, claude-desktop, mcp-server, ai-agents, retrieval-augmented-generation, nodejs, docker

## Member repositories
- CaviraOSS/OpenMemory (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:50.876840+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:38.419387+00:00, confidence not recorded.
  - readme: https://github.com/CaviraOSS/OpenMemory (fetched 2026-08-28T04:08:50.876840+00:00, sha 1c81b615cb73)
  - homepage: https://openmemory.cavira.app (fetched 2026-08-29T09:07:36.437277+00:00, sha c8da4059a142)
  - site_page: https://openmemory.cavira.app/docs/introduction (fetched 2026-08-29T09:07:36.446488+00:00, sha 2920a115e1c2)
  - site_page: https://openmemory.cavira.app/docs/quick-start (fetched 2026-08-29T09:07:36.448273+00:00, sha a6a6f9b6d5d1)
  - site_page: https://openmemory.cavira.app/docs/api/ingestion (fetched 2026-08-29T09:07:36.450211+00:00, sha f90482645a70)
  - site_page: https://openmemory.cavira.app/docs/advanced/embedding-modes (fetched 2026-08-29T09:07:36.452022+00:00, sha 810063cd39ae)
- Data as of 2026-08-30T08:39:29.467469+00:00.