Fangzhou AiandClaude Opus 5.5 3059155b47 [ROCm][Perf] MXFP8 GEMM on native 32x32 block scales for gfx950 (#58510) ...

Signed-off-by: fai <fangzhouai@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

2026-09-24 21:26:30 -05:00

..

cute_dsl

[Kernel][DSV4.1] Fuse MXFP8 wo_b GEMM with sequence-parallel reduce-scatter (#57428)

2026-09-22 01:07:17 -07:00

mixed_precision

[Quantization] Enable humming wNaM asymmetric quant (zero_point) with compressed-tensors (#46528)

2026-09-23 10:23:11 -04:00

mxfp4

[Kernel] Add opt-in load-time MXFP4 dequantization (#50814)

2026-09-22 14:51:12 +08:00

mxfp6

[Feature][Humming] Humming feature integration (#56685)

2026-09-20 08:11:01 -07:00

mxfp8

[ROCm][Perf] MXFP8 GEMM on native 32x32 block scales for gfx950 (#58510)

2026-09-24 21:26:30 -05:00

nvfp4

[Feature][Humming] Humming feature integration (#56685)

2026-09-20 08:11:01 -07:00

scaled_mm

[Feature][Humming] Humming feature integration (#56685)

2026-09-20 08:11:01 -07:00

__init__.py

[Kernel] Remove AllSpark INT8 W8A16 GEMM backend (#58001)

2026-09-22 14:07:34 -07:00

base.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

zentorch_utils.py

[CPU][Zen] Route BF16 MoE inference through zentorch on AMD (#44201)

2026-08-10 23:20:05 -07:00