Skip to content

qwen27 CUDA: a pre-sm_90 decode matmul, ~52% faster serving on an RTX 3090 (exact; stacked on #3) - #4

Open
CanadaApollo6 wants to merge 6 commits into
machinegenieorg:see-4347-qwen27-cudafrom
CanadaApollo6:sm86-qmm-lane
Open

qwen27 CUDA: a pre-sm_90 decode matmul, ~52% faster serving on an RTX 3090 (exact; stacked on #3)#4
CanadaApollo6 wants to merge 6 commits into
machinegenieorg:see-4347-qwen27-cudafrom
CanadaApollo6:sm86-qmm-lane

Commits

  1. Commits on Oct 13, 2026