Yanzhao Wang 1e7bcf3da4 metal : add FA kernels for HSK=96, HSV=64 (MiniCPM3) (#28599) ...

* metal : add FA kernels for HSK=96, HSV=64 (MiniCPM3)

MiniCPM3 sets attention.key_length to 96 and does not set
attention.value_length, which defaults to n_embd / n_head = 64. Metal had no
(96, 64) instantiation, so -fa auto aborted on the missing
kernel_flash_attn_ext_vec_f16_dk96_dv64.

Instantiate the tile kernel at (96, 64) for every K/V type that already has
(96, 96), and the vec kernel for the NE=4 configurations. Of the NE values the
vec dispatch considers, only NE=4 works here, because NL = 32/NE has to divide
both DK/4 = 24 and DV/4 = 16.

* tests : avoid redundant FA vec slice coverage

2026-09-15 11:10:40 +03:00

..

cmake

CUDA: replace GGML_FA_ALL_QUANTS with GGML_FA_QUANTS, more control over what is compiled (#28079)

2026-09-09 12:50:08 +02:00

include

ggml : update ggml_prec specification (#26675)

2026-09-08 09:06:24 +03:00

src

metal : add FA kernels for HSK=96, HSV=64 (MiniCPM3) (#28599)

2026-09-15 11:10:40 +03:00

.gitignore

vulkan : cmake integration (#8119)

2024-07-13 18:12:39 +02:00

CMakeLists.txt

ggml : bump version to 0.24.0 (ggml/1627)

2026-09-14 16:45:33 +03:00