Skip to content

[gemma4_31b][cuda] length-aware bf16 global attention#20506

Open
Gasoonjia wants to merge 1 commit into
gemma4_31b-cuda-decode-speedupfrom
gemma4_31b-cuda-attn-perf-git
Open

[gemma4_31b][cuda] length-aware bf16 global attention#20506
Gasoonjia wants to merge 1 commit into
gemma4_31b-cuda-decode-speedupfrom
gemma4_31b-cuda-attn-perf-git

[gemma4_31b][cuda] length-aware bf16 global attention + head_dim-agno…

ce442fe
Select commit
Loading
Failed to load commit list.

Select a check to view from the sidebar