Skip to content

cuda window_attention: head dim 256, any GQA, a row's last key and fp32 probabilities - #1123

Closed
jschmied wants to merge 1 commit into
ashhart:mainfrom
jschmied:cuda-window-gemma
Closed

cuda window_attention: head dim 256, any GQA, a row's last key and fp32 probabilities#1123
jschmied wants to merge 1 commit into
ashhart:mainfrom
jschmied:cuda-window-gemma

cuda window_attention: head dim 256, any GQA, a row's last key and fp…

a2bed92
Select commit
Loading
Failed to load commit list.
Sign in for the full log view

Annotations

1 warning and 1 notice
zig
succeeded Oct 20, 2026 in 9m 59s