Skip to content

cuda: native PTQ1_0 MMQ tile loader (closes the dequantize-to-cuBLAS prefill fallback) - #160

Merged
bri-prism merged 1 commit into
prismfrom
perf/ptq1_0-cuda-mmq-prism
Sep 8, 2026
Merged

cuda: native PTQ1_0 MMQ tile loader (closes the dequantize-to-cuBLAS prefill fallback)#160
bri-prism merged 1 commit into
prismfrom
perf/ptq1_0-cuda-mmq-prism

cuda: native PTQ1_0 MMQ tile loader and packed dequantize

d14a927
Select commit
Loading
Failed to load commit list.
Sign in for the full log view

Annotations

2 errors
Job log options

This job was cancelled