George HotzandGitHub 61e104bdfb use UOp.wmma everywhere (#17045) ...

* use UOp.wmma everywhere

* fix

2026-07-16 10:40:48 -07:00

..

amd

fused qkv rope custom kernel (#17021)

2026-07-15 01:08:42 +09:00

cuda

tk fa: use 16x64 tiles (#13086)

2025-11-03 18:25:38 -08:00

metal

feat: add thunderkittens (#12590)

2025-10-10 00:32:33 -07:00

tiny

use UOp.wmma everywhere (#17045)

2026-07-16 10:40:48 -07:00