Skip to content

fuse fp8 quant in kv copying and add flashinfer decode mla operator in the attention module#737

Merged
hiworldwzj merged 4 commits intoModelTC:mainfrom
blueswhen:mla_fp8
Feb 26, 2025
Merged

fuse fp8 quant in kv copying and add flashinfer decode mla operator in the attention module#737
hiworldwzj merged 4 commits intoModelTC:mainfrom
blueswhen:mla_fp8

Commits

Commits on Feb 26, 2025