Skip to content

fix(cuda/attention): bound long-prompt prefill memory for flash-ineligible models - #676

Merged
inureyes merged 3 commits into
mainfrom
fix/672-cuda-longprompt-chunked-attention
Jul 6, 2026
Merged

fix(cuda/attention): bound long-prompt prefill memory for flash-ineligible models#676
inureyes merged 3 commits into
mainfrom
fix/672-cuda-longprompt-chunked-attention

chore(benchmarks): add post-fix long-prompt sweeps for #648 and #672

31695fa
Select commit
Loading
Failed to load commit list.

Select a check to view from the sidebar