]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: add bf16 and f32 support to cublas_mul_mat_batched (llama/14361)
authorAman Gupta <redacted>
Sat, 28 Jun 2025 17:30:53 +0000 (01:30 +0800)
committerGeorgi Gerganov <redacted>
Tue, 1 Jul 2025 08:52:14 +0000 (11:52 +0300)
commit67fc917af3e5b3ec6144f7a27e82deaa889c7078
tree34424453119235077f8eb38ac041decdaeeb091a
parent6948c1f455ded4e94adc3d2bcecda509004e3266
CUDA: add bf16 and f32 support to cublas_mul_mat_batched (llama/14361)

* CUDA: add bf16 and f32 support to cublas_mul_mat_batched

* Review: add type traits and make function more generic

* Review: make check more explicit, add back comments, and fix formatting

* Review: fix formatting, remove useless type conversion, fix naming for bools
src/ggml-cuda/convert.cu
src/ggml-cuda/convert.cuh
src/ggml-cuda/ggml-cuda.cu
tests/test-backend-ops.cpp