qwen-image-edit-2509
0 of 3 operations have a usable best known on NVIDIA B200 · 3 without one
Best known on NVIDIA B200
Vision rotary position embedding generation backwardsuite of 16 casesRIAC as wellnot usable3.44µsReported · license unknown · no source2026-08-21
No entry in this cohort passes the usability policy (no public source, no install recipe, license unknown); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1. 25 more entries in this cohort.
Decoder layer fused attention MLPsuite of 16 casesAmir M. Mir | SF Tensornot usable827.2µsReported · license unknown · no source2026-07-28
No entry in this cohort passes the usability policy (no public source, no install recipe, license unknown); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1. 24 more entries in this cohort.
Cu seqlens variable length vision attentionsuite of 16 casesRIAC as wellnot usable51.8µsReported · license unknown · no source2026-08-16
No entry in this cohort passes the usability policy (no public source, no install recipe, license unknown); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1. 15 more entries in this cohort.
Coverage gaps on NVIDIA B200
No eligible evidence yet for these operations. Challenges →