Jeff Bolz 485c5c3b3b vulkan: optimize mul_mat_id loading row ids into shared memory (llama/15427) ...

- Spread the work across the whole workgroup. Using more threads seems to
far outweigh the synchronization overhead.
- Specialize the code for when the division is by a power of two.

2025-09-20 13:42:40 +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

vulkan: optimize mul_mat_id loading row ids into shared memory (llama/15427)

2025-09-20 13:42:40 +03:00

.gitignore

whisper : reorganize source code + improve CMake (#2256)

2024-06-26 19:34:09 +03:00

CMakeLists.txt

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

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