]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
CUDA: skip fully masked-out KV in FA vec kernel (llama/13584)
authorJohannes Gäßler <redacted>
Tue, 20 May 2025 12:45:07 +0000 (14:45 +0200)
committerGeorgi Gerganov <redacted>
Tue, 27 May 2025 15:03:00 +0000 (18:03 +0300)
commit926fe234e92b5627b01e613fb6e9495316276fb0
treeb442817e8fd2fc24a1915a4b17e80866c0a6cceb
parentf44b53480f084d02deb42f5b96d50057005bbfb3
CUDA: skip fully masked-out KV in FA vec kernel (llama/13584)

* CUDA: skip fully masked-out KV in FA vec kernel
ggml/src/ggml-cuda/fattn-vec-f16.cuh
ggml/src/ggml-cuda/fattn-vec-f32.cuh