←back to thread

Kimi Linear: An Expressive, Efficient Attention Architecture

(github.com)
210 points blackcat201 | 1 comments | 31 Oct 25 00:07 UTC | HN request time: 0s | source
Show context
logicartisan ◴[31 Oct 25 11:58 UTC] No.45771056[source]▶
Amazing how fast AI keeps improving, every new model feels like a big step forward
replies(1): >>45771240 #
hirako2000 ◴[31 Oct 25 12:21 UTC] No.45771240[source]▶
It solely is improving on efficiency. While it is extremely valuable given the disproportionate (to value) costs of these things, your statement almost sounds like it has improved an even more challenging aspect, pushing performance.
replies(4): >>45771821 #>>45771931 #>>45774235 #>>45774316 #
1. acuozzo ◴[31 Oct 25 17:11 UTC] No.45774316[source]▶
> It solely is improving on efficiency.

Consider the implications of increases in efficiency *when you hold compute constant*.

The win is far more obvious when it's "we can do more with what we have" instead of "we can do the same with less".