Skip to content

metal: TQ1_0 kernel performance (dense decode +39-55%, prefill +78-90%) - #156

Merged
bri-prism merged 4 commits into
feat/tq1_0-cuda-v7from
perf/tq1_0-metal-flatten
Sep 8, 2026
Merged

bri-prism merged 4 commits into
feat/tq1_0-cuda-v7from
perf/tq1_0-metal-flatten

metal: TQ1_0 mat-vec, 8 rows per simdgroup

42e5d6b
Select commit
Loading
Failed to load commit list.
Sign in for the full log view