Skip to content

[ROCm] Prefill performance optimization for embedding models#1102

Open
liaocz wants to merge 1 commit into
mainfrom
feat/opt_emb
Open

[ROCm] Prefill performance optimization for embedding models#1102
liaocz wants to merge 1 commit into
mainfrom
feat/opt_emb

perf(rocm): skip fusedQKV transpose for embedding models in attention…

df5193d
Select commit
Loading
Failed to load commit list.