]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: limit number of FA stream-k CUDA blocks (llama/20586)
authorJohannes Gäßler <redacted>
Sun, 15 Mar 2026 17:30:47 +0000 (18:30 +0100)
committerGeorgi Gerganov <redacted>
Sun, 15 Mar 2026 19:50:13 +0000 (21:50 +0200)
commit6c8d7fad4f4f85456576d6888b335dce28b2c8dd
tree98eac12d8be0bf3f6134a8873304b3f1c99f078c
parent7428e41db61f3f1162016cdc89eb63e07956165b
CUDA: limit number of FA stream-k CUDA blocks (llama/20586)
src/ggml-cuda/fattn-common.cuh