Skip to content

[ROCm] Add head-fused HIP sparse MLA decode kernel for gfx950 - #57282

Open
nehaprakriya wants to merge 1 commit into
vllm-project:mainfrom
nehaprakriya:rocm/hip-m5-sparse-mla-decode
Open

nehaprakriya wants to merge 1 commit into
vllm-project:mainfrom
nehaprakriya:rocm/hip-m5-sparse-mla-decode