Michael Goin c3aea10dc8

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280) ...

Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

2025-09-11 15:43:14 -07:00

..

cutlass_gemm_caller.cuh

…

scaled_mm_azp_sm90_int8.cu

…

scaled_mm_blockwise_sm90_fp8_dispatch.cuh

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280)

2025-09-11 15:43:14 -07:00

scaled_mm_blockwise_sm90_fp8.cu

…

scaled_mm_blockwise_sm100_fp8_dispatch.cuh

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280)

2025-09-11 15:43:14 -07:00

scaled_mm_blockwise_sm100_fp8.cu

…

scaled_mm_blockwise_sm120_fp8_dispatch.cuh

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280)

2025-09-11 15:43:14 -07:00

scaled_mm_blockwise_sm120_fp8.cu

[Kernel] Add support for block FP8 on SM120 (NVIDIA 5090 and RTX PRO 6000) (#22131)

2025-08-07 19:18:28 -07:00

scaled_mm_helper.hpp

[Perf] Use upstream CUTLASS for SM90 Block FP8 kernel (#23280)

2025-09-11 15:43:14 -07:00

scaled_mm_kernels.hpp

[Kernel] Add support for block FP8 on SM120 (NVIDIA 5090 and RTX PRO 6000) (#22131)

2025-08-07 19:18:28 -07:00

scaled_mm_sm90_fp8_dispatch.cuh

…

scaled_mm_sm90_fp8.cu

…

scaled_mm_sm90_int8_dispatch.cuh

…

scaled_mm_sm90_int8.cu

…

scaled_mm_sm100_fp8_dispatch.cuh

…

scaled_mm_sm100_fp8.cu

…

scaled_mm_sm120_fp8_dispatch.cuh

…

scaled_mm_sm120_fp8.cu

…

scaled_mm.cuh

…