Oliver Simons 9bf6c3c860 CUDA: better coalesce data-access for contiguous concat (llama/22330) ... Also, distribute all elements across CTAs evenly instead of launching one CTA per dim		2026-04-30 11:29:20 +03:00
..
cmake	cmake : remove unused file (ggml/1419)	2026-02-08 09:29:10 +02:00
include	CUDA: manage NCCL communicators in context (llama/21891)	2026-04-30 11:29:09 +03:00
src	CUDA: better coalesce data-access for contiguous concat (llama/22330)	2026-04-30 11:29:20 +03:00
.gitignore	…
CMakeLists.txt	HIP: flip GGML_HIP_GRAPHS to default on (llama/22254)	2026-04-30 11:29:16 +03:00