]> git.djapps.eu Git - pkg/ggml/sources/whisper.cpp/commit
CUDA: GEMM for FP32/FP16/BF16 and ne11 <= 16 (llama/15131)
authorJohannes Gäßler <redacted>
Thu, 7 Aug 2025 08:53:21 +0000 (10:53 +0200)
committerGeorgi Gerganov <redacted>
Mon, 18 Aug 2025 17:30:45 +0000 (20:30 +0300)
commit5caf8a1ea2deb8dfbcf0778632ecb10fd8e061ba
treef015bb25dbc785b972c584474c33a1b7f09a58c5
parentb405fd88b382fd29096f34d081b338d49ebcc34a
CUDA: GEMM for FP32/FP16/BF16 and ne11 <= 16 (llama/15131)

* CUDA: GEMM for FP32/FP16/BF16 and ne11 <= 16
13 files changed:
ggml/src/ggml-cuda/common.cuh
ggml/src/ggml-cuda/fattn-mma-f16.cuh
ggml/src/ggml-cuda/fattn.cu
ggml/src/ggml-cuda/ggml-cuda.cu
ggml/src/ggml-cuda/mma.cuh
ggml/src/ggml-cuda/mmf.cu [new file with mode: 0644]
ggml/src/ggml-cuda/mmf.cuh [new file with mode: 0644]
ggml/src/ggml-cuda/mmq.cu
ggml/src/ggml-cuda/mmq.cuh
ggml/src/ggml-cuda/mmvf.cu [new file with mode: 0644]
ggml/src/ggml-cuda/mmvf.cuh [new file with mode: 0644]
ggml/src/ggml-cuda/vendors/hip.h
ggml/src/ggml-cuda/vendors/musa.h