Georgi Gerganov 1c21a850be metal : optimize FA vec for large sequences and BS <= 8 (llama/15566) ...

* metal : optmize FA vec for large heads and sequences

* metal : adjust small-batch mul mv kernels

ggml-ci

* batched-bench : fix total speed computation

ggml-ci

* cont : add comments

ggml-ci

2025-09-20 13:42:42 +03:00

..

cmake

ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (llama/15094)

2025-08-18 20:30:45 +03:00

include

ggml: add conv3d op (llama/15182)

2025-09-20 13:42:39 +03:00

src

metal : optimize FA vec for large sequences and BS <= 8 (llama/15566)

2025-09-20 13:42:42 +03:00

.gitignore

…

CMakeLists.txt

CUDA: replace GGML_CUDA_F16 with CUDA arch checks (llama/15433)

2025-09-20 13:42:38 +03:00