llama-3-2-3b · KernelIndex

0 of 12 operations have a usable best known on NVIDIA GB200 · 12 without one

Best known on NVIDIA GB200

Operation · workload

Best known

Record

Trust

Observed

gqa-paged-attention2 operations

GQA paged prefill causal h24 kv8 d128 ps1bf16 · [1, 24, 128]

flashinfer / wrapperb38b5fnot usable

87.8µs

Reported · Apache-2.0 · source

2026-04-12

No entry in this cohort passes the usability policy (no install recipe); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1.

source mirroredApache-2.0no install recipeCohort on the operation page →View source →Vendor the source →Run detail →

GQA paged prefill causal h24 kv8 d128 ps64bf16 · [1, 24, 128]

flashinfer / wrapper92685dnot usable

39.0µs

Reported · Apache-2.0 · source

2026-04-12

No entry in this cohort passes the usability policy (no install recipe); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1.

source mirroredApache-2.0no install recipeCohort on the operation page →View source →Vendor the source →Run detail →

gqa-ragged-attention1 operation

GQA ragged prefill causal h24 kv8 d128bf16 · [1, 8, 128]

flashinfer / wrapperfe59e7not usable

41.4µs

Reported · Apache-2.0 · source

2026-04-12

No entry in this cohort passes the usability policy (no install recipe); the fastest known is shown. Measured by the source's own harness. Ranked by its metric under ranking-v1.

source mirroredApache-2.0no install recipeCohort on the operation page →View source →Vendor the source →Run detail →