]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
overview / pkg / ggml / sources / whisper.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: e89fdce)
CUDA: revert part of the RDNA1 optimizations (llama/8309)
authorDaniele <redacted>
Fri, 5 Jul 2024 07:06:09 +0000 (07:06 +0000)
committerGeorgi Gerganov <redacted>
Mon, 8 Jul 2024 11:53:55 +0000 (14:53 +0300)
commit73703a144fd9d14c104932813898352549acd817
tree718e287ce90e275150fe4ec950ddfa6cba162013tree
parente89fdceec218fb94e7fd3afc184826b168ab7406commit | diff
CUDA: revert part of the RDNA1 optimizations (llama/8309)

The change on the launch_bounds was causing a small performance drop in perplexity of 25 t/s
ggml/src/ggml-cuda/mmq.cuh diff | blob | history
Packaging of ggerganov/whisper.cpp