Sign in Subscribe

vecq

Head-to-head comparison of vecq, usearch, and qdrant-edge vector engines: 5x smaller vectors headline

vecq vs usearch vs qdrant-edge: one RAM budget, three different answers

Dark thumbnail with glowing purple data blocks on the right; headline reads 4-bit on the edge: what holds, what gives

4-Bit Vectors on the Edge: Where Recall Holds and Where It Gives

Purple quantization wave art behind the headline: half a million tokens of context still needs an index

Half a Million Tokens of Context Still Needs an Index

Violet data blocks compressing into a dense stack on a dark background. Headline: Your RAG stack got lean. Your vectors did not.

Your RAG stack got lean. Your vector layer didn't. Meet vecq.

Headline graphic: 67 cents of compute, 44% on ARC-AGI-1, small transformer trained in 1.5 hours beating many LLMs.

A Small Transformer Trained in 1.5 Hours Beat Many LLMs on ARC

Dark abstract purple vector ribbons on black with the headline: 87GB on a CPU. Quantization did that.

An 87GB model on a CPU: what the quantization math buys you