Skip to content
KernelIndex
Search⌘K
Sign in

nova-70b-nvfp4

best known per operation on the selected GPU
3 operations · 1 family · 66 eligible runs
GPUNVIDIA B20066

0 of 3 operations have a usable best known on NVIDIA B200 · 3 without one

Best known on NVIDIA B200

Operation · workload
Best known
Record
Trust
Observed
quantization3 operations
NVFP4 attention output projection with residualsuite of 16 cases
Geometricnot usable
66.5µs
Reported · license unknown · no source
2026-08-14

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. 16 more entries in this cohort.

no sourcelicense unknownno install recipeCohort on the operation page →Run detail →
NVFP4 grouped query attentionsuite of 16 cases
Lasernot usable
540.5µs
Reported · license unknown · no source
2026-08-23

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. 16 more entries in this cohort.

no sourcelicense unknownno install recipeCohort on the operation page →Run detail →
NVFP4 linear layersuite of 16 cases
Amir M. Mir | SF Tensornot usable
87.2µs
Reported · license unknown · no source
2026-08-20

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. 13 more entries in this cohort.

no sourcelicense unknownno install recipeCohort on the operation page →Run detail →

Coverage gaps on NVIDIA B200

Operation
Family
Status
NVFP4 attention output projection with residual
quantization
measured · no usable entry (no public source, no install recipe, license unknown)
NVFP4 grouped query attention
quantization
measured · no usable entry (no public source, no install recipe, license unknown)
NVFP4 linear layer
quantization
measured · no usable entry (no public source, no install recipe, license unknown)

No eligible evidence yet for these operations. Challenges →

Sources: NVIDIA SOL-ExecBench (2026-08-23)How comparison worksJSON