Launched this week
HMLR

HMLR

Living memory system for AI agents

2 followers

Visit website
HMLR (Hierarchical Memory Lookup & Routing) is an open-source, state-aware long-term memory architecture that gives AI agents true "living" memory but without massive context windows or fragile RAG. It delivers verified multi-hop reasoning, temporal conflict resolution, and cross-topic persistence Ideal for building reliable agentic AI that remembers forever, cheaply. Feedback, stars, and contributions welcome!
HMLR gallery image
Free
Launch tags:Open Source•Developer Tools•Artificial Intelligence
Launch Team / Built With
Intercom
Intercom
Startups get 90% off Intercom + 1 year of Fin AI Agent free
Promoted

What do you think? …

Sean VanWinkle
Maker
📌
Hi Product Hunt! I'm Sean, the maker behind HMLR, an open-source memory system designed to give AI agents persistent, reliable long-term memory, and also completly free! Tired of agents forgetting facts across conversations or failing on multi-hop/policy tests? HMLR solves that with hierarchical routing, fact scrubbing, and validated retrieval, achieving perfect scores on benchmarks that stump most systems. It's MIT-licensed, pip installable (pip install hmlr), and integrates easily with LangGraph. Would love your feedback, stars, or ideas for improvements! Try the simple_agent example and let me know what you build. 🚀 Thanks for checking it out!