1.
2.
Querying 3B Vectors
(news.ycombinator.com)
3.
Connes Embedding Problem
(news.ycombinator.com)
4.
5.
ChromaDB Explorer
(news.ycombinator.com)
6.
Using Vectorize to build an unreasonably good search engine in 160 lines of code
(news.ycombinator.com)
7.
Prompt caching for cheaper LLM tokens
(news.ycombinator.com)
8.
Prompt caching: 10x cheaper LLM tokens, but how?
(news.ycombinator.com)
9.
Contextualization Machines
(news.ycombinator.com)
10.
28M Hacker News comments as vector embedding search dataset
(news.ycombinator.com)
11.
Why Google’s File Search could displace DIY RAG stacks in the enterprise
(venturebeat.com)
12.
Ilo – a Forth system running on UEFI
(news.ycombinator.com)
13.
Meta Superintelligence Labs' first paper is about RAG
(news.ycombinator.com)
14.
Meta Superintelligence's surprising first paper
(news.ycombinator.com)
15.
Language models pack billions of concepts into 12k dimensions
(news.ycombinator.com)
16.
Language Models Pack Billions of Concepts into 12k Dimensions
(news.ycombinator.com)
17.
Language Models Pack Billions of Concepts into 12,000 Dimensions
(news.ycombinator.com)
18.
How big are our embeddings now and why?
(news.ycombinator.com)
19.
We Hit 100% GPU Utilization–and Then Made It 3× Faster by Not Using It
(news.ycombinator.com)
20.
Show HN: Bolt – A super-fast, statically-typed scripting language written in C
(news.ycombinator.com)
21.
Gemini Embedding: Powering RAG and context engineering
(news.ycombinator.com)
22.
23.
All AI models might be the same
(news.ycombinator.com)
24.
All AI Models Might be The Same
(news.ycombinator.com)
25.
LGND wants to make ChatGPT for the Earth
(techcrunch.com)
26.
Muvera: Making multi-vector retrieval as fast as single-vector search
(news.ycombinator.com)