Commit 005756a

committed

vulkan: scalar flash attention implementation

1 parent 9070365 commit 005756aCopy full SHA for 005756a

3 files changed

+535

-76

lines changed

ggml/src/ggml-vulkan
- ggml-vulkan.cpp
- vulkan-shaders
  - flash_attn.comp
  - vulkan-shaders-gen.cpp

3 files changed

+535

-76

lines changed

Comments

(0)