JasonDocton/lucid-memory
Memory for AI that works like yours—local, instant, persistent. 13x faster than Pinecone, 5x leaner than RAG. Finds what RAG misses. Zero cloud, zero cost.
Uses context-aware spreading activation and multiplicative recency weighting to surface related memories while preserving relevance—achieving 82.5% recall versus 55% for vector databases. Implements gist compression (~37 tokens per memory vs ~75) and a Rust core that processes 743,000 memories/second locally, eliminating cloud latency and costs. Integrates natively with Claude Code, OpenAI Codex, and OpenCode through MCP with automatic memory capture via hooks and plugins.
132 stars.
Stars
132
Forks
7
Language
TypeScript
License
GPL-3.0
Category
Last pushed
Feb 20, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/JasonDocton/lucid-memory"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
topoteretes/cognee
Knowledge Engine for AI Agent Memory in 6 lines of code
divagr18/memlayer
Plug-and-play memory for LLMs in 3 lines of code. Add persistent, intelligent, human-like memory...
verygoodplugins/automem
AutoMem is a graph-vector memory service that gives AI assistants durable, relational memory:
CortexReach/memory-lancedb-pro
Enhanced LanceDB memory plugin for OpenClaw — Hybrid Retrieval (Vector + BM25), Cross-Encoder...
CaviraOSS/OpenMemory
Local persistent memory store for LLM applications including claude desktop, github copilot,...