]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: faster Deepseek FA, add Turing support (llama/13435)
authorJohannes Gäßler <redacted>
Wed, 14 May 2025 14:08:20 +0000 (16:08 +0200)
committerGeorgi Gerganov <redacted>
Mon, 19 May 2025 10:37:56 +0000 (13:37 +0300)
commitc91952e75b208479622a395456307dc40cd7c827
treefed579382c47caa7a6424a7cacf372aef25f747c
parent84f4641e8f08050b5b1309b184be82bddd39464a
CUDA: faster Deepseek FA, add Turing support (llama/13435)
src/ggml-cuda/fattn-common.cuh
src/ggml-cuda/fattn-mma-f16.cuh
src/ggml-cuda/fattn.cu
src/ggml-cuda/ggml-cuda.cu