Mnemosyne Documentation
June 14, 2026 ยท View on GitHub
Local-first, zero-cloud memory for AI agents. SQLite-backed. Sub-millisecond. Fully private.
Guides
| Document | Description |
|---|---|
| Getting Started | Installation, quickstart, storing your first memory |
| Architecture | BEAM tiers, SQLite backend, hybrid search, knowledge graph |
| API Reference | Python API: remember, recall, sleep, triples, stats |
| Integrations | Platform guides: Cursor, Claude Code, Codex, OpenWebUI, Windsurf, OpenClaw, Hermes |
| OpenWebUI Deep Integration | Auto-save every chat, memory browser dashboard, cross-session recall |
| Integration Template | ~100-line pattern for adding any new platform |
| Hermes Integration | Using Mnemosyne as a Hermes memory backend |
| LLM Installation Guide | Installation instructions for AI agents/LLMs |
| Configuration | Environment variables, data directory, vector compression |
| Mnemosyne Sync | Bidirectional sync, encryption, deployment, tutorial |
| Security & Privacy | Threat model, encryption internals, BYOK comparison |
| Benchmarking | Maintainer guide: per-tool A/B benchmark env vars, diagnostics, pure-recall mode, test sequence template |
| Benchmark Results Analysis | Output-file schemas + analysis recipes (per-ability scores, paired bootstrap CIs, voice attribution). AI-assistant-friendly reference |
| Changelog | Version history and release notes |
Quick Links
- Repository: github.com/AxDSan/mnemosyne
- PyPI: mnemosyne-memory
- Contributing: See CONTRIBUTING.md in the repo root
- Updating: See UPDATING.md for upgrade and rollback instructions
- License: MIT -- see LICENSE