| .. |
|
cmake
|
9a96389544
ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094)
|
5 months ago |
|
include
|
ef07a40906
ggml webgpu: add support for soft_max, optimize rms_norm (#16357)
|
3 months ago |
|
src
|
e308efda8e
vulkan: in flash attention, bounds check against nem1 (don't rely on GGML_KQ_MASK_PAD) (#16316)
|
3 months ago |
|
.gitignore
|
17eb6aa8a9
vulkan : cmake integration (#8119)
|
1 year ago |
|
CMakeLists.txt
|
e95fec640f
HIP: Disable ROCWMMA fattn on CDNA when compiled against ROCWMMA 2.0.0 (#16221)
|
3 months ago |
Jeff Bolz
e308efda8e
vulkan: in flash attention, bounds check against nem1 (don't rely on GGML_KQ_MASK_PAD) (#16316)