Kawrakow 555330fbba CUDA: handle GQA = 12 for head size = 256 via new MMA (#2372) ... * iCUDA: handle GQA = 12 for head size = 256 via new MMA * Cleanup		2026-08-28 18:10:50 +02:00
..
cmake	…
include	Quantization fudge factors (#2361)	2026-08-27 17:35:04 +02:00
src	CUDA: handle GQA = 12 for head size = 256 via new MMA (#2372)	2026-08-28 18:10:50 +02:00
.gitignore	…
CMakeLists.txt	Chunked experts (CPU) (#2202)	2026-07-30 13:16:02 +03:00