]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: fix MMQ nwarps for AMD with warp_size==32 (llama/15014)
authorJohannes Gäßler <redacted>
Fri, 1 Aug 2025 18:47:32 +0000 (20:47 +0200)
committerGeorgi Gerganov <redacted>
Sat, 2 Aug 2025 14:51:21 +0000 (17:51 +0300)
commit5b41763a4ec2c76ada457b7ee97450a028c44235
tree165cc9f8d2e8e4fd762a5fd77821f9d12f3ce94a
parent6a06b78fb969e58a5614fb87c798252564370a36
CUDA: fix MMQ nwarps for AMD with warp_size==32 (llama/15014)
src/ggml-cuda/mmq.cuh