Stores knowledge as Smart Frames (content + embedding + time + relationships)Hybrid search with sub-5ms retrievalFully portable and open-sourceReplaces RAG + vector DBs entirely with a single portable memory file

Not

Hacker

News!

AI-observed conversations & context

Daily AI-observed summaries, trends, and audience signals pulled from Hacker News so you can see the conversation before it hits your feed.

LiveBeta

Explore

Home
Hiring
Products
Companies
Discussion
Q&A

Tech Stack

GPTClaudeGeminiLlamaLangChainAutogen

memory localityRAG pipelinesvector databasesMachine LearningArtificial Intelligence

Discussion (2 comments)

Showing 2 comments

ifh-hn

3d ago

I have no experience with video encoding, or AI training. My only interactions with either are as a consumer.

I have no idea how video relates to AI memory and I'm more confused after reading the above and the GitHub.

All that being said, my understanding is this is model independent single file bespoke AI memory that isn't a bunch of different technologies taped together. Meaning less overhead and complexity.

If that's correct then kudos, well done on the insight, even if by accident.

I just wish I understood any of it.

mohamed_9724

4d ago

End of an era!

View on Hacker NewsHN ID: 46502628Mood: excited

Resources

Visit Hacker News
HN API
Modal cronjobs
Meta Llama

Briefings

Inbox recaps on the loudest debates & under-the-radar launches.

Connect

Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.