Xiaoyu Zhang 2c8fd99363 [sgl-kernel] per token group quant support COLUMN MAJOR (#4817)

2025-04-02 18:29:59 -07:00

..

allreduce

fix custom allreduce performance/accuracy problem (#4477)

2025-03-16 12:16:30 -07:00

attention

support cmake for sgl-kernel (#4706)

2025-03-27 01:42:28 -07:00

cutlass_extensions

Rename files in sgl kernel to avoid nested folder structure (#4213)

2025-03-08 22:54:51 -08:00

elementwise

Optimize rope in sgl kernel (#4267)

2025-03-10 10:07:45 -07:00

gemm

[sgl-kernel] per token group quant support COLUMN MAJOR (#4817)

2025-04-02 18:29:59 -07:00

moe

Add deepseek style fused moe group gate selection kernel (#4530)

2025-03-29 11:51:45 -07:00

speculative

[ROCm] Enable MTP (NextN) on AMD GPU (#4631)

2025-03-23 22:58:05 -07:00

torch_extension_rocm.cc

[ROCm] Enable MTP (NextN) on AMD GPU (#4631)

2025-03-23 22:58:05 -07:00

torch_extension.cc

[Fix] revert clean m.def for cudagraph (#4944)

2025-03-31 02:08:55 -07:00