Skip to content

[Kernel][Attention] Separate Attention.kv_scale into k_scale and v_scale#6081

Merged
simon-mo merged 13 commits intovllm-project:mainfrom neuralmagic:separate-key-value-scalesJul 16, 2024

Commits

Commits on Jul 3, 2024

Commits on Jul 15, 2024

Commits on Jul 16, 2024