Skip to content

[Core] Optimizing cross-attention QKVParallelLinear computation#12325

Merged
DarkLight1337 merged 10 commits intovllm-project:mainfrom
NickLucche:encdec-separate-crossattn
Mar 6, 2025
Merged

[Core] Optimizing cross-attention `QKVParallelLinear` computation#12325
DarkLight1337 merged 10 commits intovllm-project:mainfrom
NickLucche:encdec-separate-crossattn

Commits

Commits on Mar 5, 2025