qwen3-8b
best known per operation on the selected GPU
5 operations · 3 families · 91 eligible runs
0 of 5 operations have a usable best known on NVIDIA B200 · 5 without one
Best known on NVIDIA B200
Operation · workload
Best known
Record
Trust
Observed
rope1 operation
RoPE inverse frequency computationsuite of 16 casesac4knot usable896nsReported · license unknown · no source2026-07-10
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. 30 more entries in this cohort.
no sourcelicense unknownno install recipeCohort on the operation page →Run detail →
rmsnorm1 operation
Grouped query attention with RoPE and QK normsuite of 16 casesAmir M. Mir | SF Tensornot usable228.4µsReported · license unknown · no source2026-08-07
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.
no sourcelicense unknownno install recipeCohort on the operation page →Run detail →
Coverage gaps on NVIDIA B200
Operation
Family
Status
rope
measured · no usable entry (no public source, no install recipe, license unknown)
rmsnorm
measured · no usable entry (no public source, no install recipe, license unknown)
No eligible evidence yet for these operations. Challenges →
Related tags: qwen3-8b-nvfp4
Sources: NVIDIA SOL-ExecBench (2026-08-24)How comparison worksJSON