Ximingwang-09 2dd6489468 Add H20 dtype fp8_w8a8 shared experts fused MoE kernel tuning configs for DeepSeek V3/R1 (#5291) ...

Co-authored-by: ximing.wxm <ximing.wxm@antgroup.com>

2025-04-14 18:40:31 -07:00

..

ep_moe

[DeepEP] fix: import buffer error (#5179)

2025-04-08 22:12:14 -07:00

fused_moe_triton

Add H20 dtype fp8_w8a8 shared experts fused MoE kernel tuning configs for DeepSeek V3/R1 (#5291)

2025-04-14 18:40:31 -07:00

fused_moe_native.py

Add Llama4 support (#5092)

2025-04-07 00:29:36 -07:00

router.py

Fix run time error in ROCm platform (#5147)

2025-04-07 22:49:40 -07:00

topk.py

apply fused moe gate in ds v3/r1 (#5371)

2025-04-14 16:24:26 -07:00