3 people 2286e85e77 pass a_scale from fp8 quant result instead of hard code to 1.0f (#10241) ...

Co-authored-by: Yichen Wang <yichen.wang@bytedance.com>
Co-authored-by: Jinwu Guo <641876696@qq.com>

2025-09-10 12:56:05 -07:00

..

scaled_mm_entry.cu

[1/n]: add cutlass W4A8 moe kernel for hopper architecture (#7772)

2025-07-04 20:50:12 -07:00

w4a8_get_group_starts.cuh

[feat] Support tp mode for DeepSeek-R1-W4AFP8 (#8118)

2025-09-01 22:17:26 -07:00

w4a8_grouped_mm_c3x.cu

[feat] Support tp mode for DeepSeek-R1-W4AFP8 (#8118)

2025-09-01 22:17:26 -07:00

w4a8_grouped_mm_c3x.cuh

pass a_scale from fp8 quant result instead of hard code to 1.0f (#10241)

2025-09-10 12:56:05 -07:00

w4a8_moe_data.cu

[1/n]: add cutlass W4A8 moe kernel for hopper architecture (#7772)

2025-07-04 20:50:12 -07:00