Hongqiang Wang 9f31776c37 opencl: choose the MoE expert matmul by batch size for speculative decoding/MTP (#27637) ...

* opencl: gate the prebuilt q4_0 MoE GEMM on routing count

* opencl: stop writing zeros into the padded MoE activation slots

* opencl: rephrase claude's comments

---------

Co-authored-by: Li He <lih@qti.qualcomm.com>

2026-09-15 11:21:05 -07:00

..

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

rpc : hash-cache only weights (#28789)

2026-09-15 14:50:20 +03:00

src

opencl: choose the MoE expert matmul by batch size for speculative decoding/MTP (#27637)

2026-09-15 11:21:05 -07:00

.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