Skip to content
KernelIndex
Search⌘K

GEMM n5120 k2048

208 eligible runs
gemm

General matrix multiply (GEMM) C = A @ B.T. Captured from Qwen 3 30B A3B attn.qkv_proj.

Fastest reported
18.6µsmean · 51% of SOL
Amir M. Mir | SF TensorLicense unknown · unknown

Reported evidence · last observed 2026-08-02. Reported by source; not independently reproduced. No public source. License unknown.

No public sourceRun detail →

Current records

Source-native comparison · GPU NVIDIA B200 · Workload suite of 25 cases · mean latency · Protocol SOL-ExecBench evaluation stack v1.1 · mean · 43 results · last observed 2026-08-21Record history →
#
Implementation
Latency
vs #1
Trust
Observed
1
18.6µs51% SOL
1.00×
Reported · license unknown · no source
2026-08-02

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
2
19.0µs53% SOL
1.02×
Reported · license unknown · no source
2026-08-04

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
3
19.1µs53% SOL
1.02×
Reported · license unknown · no source
2026-08-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
4
19.2µs53% SOL
1.03×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
5
19.2µs52% SOL
1.03×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
6
19.2µs50% SOL
1.03×
Reported · license unknown · no source
2026-08-02

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
7
19.5µs52% SOL
1.04×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
8
19.5µs50% SOL
1.05×
Reported · license unknown · no source
2026-07-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
9
19.5µs50% SOL
1.05×
Reported · license unknown · no source
2026-08-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
10
19.5µs52% SOL
1.05×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
11
19.7µs50% SOL
1.06×
Reported · license unknown · no source
2026-08-05

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
12
19.7µs49% SOL
1.06×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
13
19.7µs49% SOL
1.06×
Reported · license unknown · no source
2026-08-01

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
14
19.7µs49% SOL
1.06×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
15
19.7µs49% SOL
1.06×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
16
19.8µs49% SOL
1.06×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
17
20.0µs48% SOL
1.07×
Reported · license unknown · no source
2026-08-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
18
20.0µs49% SOL
1.07×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
19
20.0µs49% SOL
1.08×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
20
20.1µs49% SOL
1.08×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
21
20.1µs49% SOL
1.08×
Reported · license unknown · no source
2026-07-25

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
22
20.1µs49% SOL
1.08×
Reported · license unknown · no source
2026-07-31

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
23
20.1µs48% SOL
1.08×
Reported · license unknown · no source
2026-08-02

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
24
20.1µs48% SOL
1.08×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
25
20.2µs49% SOL
1.09×
Reported · license unknown · no source
2026-07-23

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
26
20.3µs48% SOL
1.09×
Reported · license unknown · no source
2026-08-01

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
27
20.3µs48% SOL
1.09×
Reported · license unknown · no source
2026-07-30

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
28
20.6µs47% SOL
1.10×
Reported · license unknown · no source
2026-07-18

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
29
20.6µs47% SOL
1.11×
Reported · license unknown · no source
2026-07-23

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
30
20.7µs48% SOL
1.11×
Reported · license unknown · no source
2026-08-01

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
31
20.8µs47% SOL
1.11×
Reported · license unknown · no source
2026-07-28

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
32
20.8µs47% SOL
1.12×
Reported · license unknown · no source
2026-07-24

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
33
20.8µs47% SOL
1.12×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
34
21.0µs46% SOL
1.13×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
35
21.1µs46% SOL
1.13×
Reported · license unknown · no source
2026-07-29

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
36
21.1µs46% SOL
1.13×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
37
21.1µs46% SOL
1.13×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
38
21.1µs46% SOL
1.13×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
39
24.5µs41% SOL
1.32×
Reported · license unknown · no source
2026-07-23

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
40
25.1µs39% SOL
1.35×
Reported · license unknown · no source
2026-07-22

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
41
25.4µs38% SOL
1.36×
Reported · license unknown · no source
2026-07-10

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
42
26.8µs36% SOL
1.44×
Reported · license unknown · no source
2026-07-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →
43
27.0µs36% SOL
1.45×
Reported · license unknown · no source
2026-07-21

Measured exactly what you asked. Reported by source; not independently reproduced.

no sourcelicense unknownno install recipeRun detail →

Implementations

Implementation
Runtime
Best latency
Evidence
Availability
unknown
19.7µs
1.64×
Reported
license unknown · no source
unknown
19.0µs
1.58×
Reported
license unknown · no source
unknown
19.2µs
1.60×
Reported
license unknown · no source
unknown
19.5µs
1.62×
Reported
license unknown · no source
unknown
19.2µs
1.59×
Reported
license unknown · no source
unknown
19.2µs
1.60×
Reported
license unknown · no source
unknown
18.6µs
1.55×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
Show all 67 implementations ›
unknown
19.8µs
1.64×
Reported
license unknown · no source
unknown
20.2µs
1.68×
Reported
license unknown · no source
unknown
19.9µs
1.65×
Reported
license unknown · no source
unknown
19.5µs
1.62×
Reported
license unknown · no source
unknown
20.6µs
1.71×
Reported
license unknown · no source
unknown
20.8µs
1.73×
Reported
license unknown · no source
unknown
21.1µs
1.75×
Reported
license unknown · no source
unknown
21.1µs
1.75×
Reported
license unknown · no source
unknown
22.3µs
1.86×
Reported
license unknown · no source
unknown
25.1µs
2.08×
Reported
license unknown · no source
unknown
19.1µs
1.59×
Reported
license unknown · no source
unknown
19.5µs
1.62×
Reported
license unknown · no source
unknown
19.2µs
1.60×
Reported
license unknown · no source
unknown
19.5µs
1.62×
Reported
license unknown · no source
unknown
19.7µs
1.63×
Reported
license unknown · no source
unknown
19.9µs
1.65×
Reported
license unknown · no source
unknown
19.7µs
1.64×
Reported
license unknown · no source
unknown
19.7µs
1.64×
Reported
license unknown · no source
unknown
19.7µs
1.64×
Reported
license unknown · no source
unknown
19.8µs
1.64×
Reported
license unknown · no source
unknown
20.0µs
1.66×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
unknown
20.0µs
1.67×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
unknown
20.2µs
1.68×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
unknown
20.1µs
1.67×
Reported
license unknown · no source
unknown
20.3µs
1.69×
Reported
license unknown · no source
unknown
20.0µs
1.66×
Reported
license unknown · no source
unknown
20.7µs
1.72×
Reported
license unknown · no source
unknown
20.3µs
1.69×
Reported
license unknown · no source
unknown
20.8µs
1.73×
Reported
license unknown · no source
unknown
20.6µs
1.72×
Reported
license unknown · no source
unknown
20.8µs
1.73×
Reported
license unknown · no source
unknown
21.1µs
1.76×
Reported
license unknown · no source
unknown
21.1µs
1.75×
Reported
license unknown · no source
unknown
21.0µs
1.74×
Reported
license unknown · no source
unknown
21.1µs
1.75×
Reported
license unknown · no source
unknown
22.7µs
1.89×
Reported
license unknown · no source
unknown
24.0µs
1.99×
Reported
license unknown · no source
unknown
24.0µs
1.99×
Reported
license unknown · no source
unknown
24.0µs
2.00×
Reported
license unknown · no source
unknown
24.5µs
2.04×
Reported
license unknown · no source
unknown
24.8µs
2.06×
Reported
license unknown · no source
unknown
24.8µs
2.06×
Reported
license unknown · no source
unknown
25.4µs
2.11×
Reported
license unknown · no source
unknown
26.8µs
2.23×
Reported
license unknown · no source
unknown
27.0µs
2.24×
Reported
license unknown · no source
unknown
28.0µs
2.32×
Reported
license unknown · no source
cuda
—
—
Apache-2.0 · source
triton
26.5µs
2.20×
Reported
Apache-2.0 · source
cuda
—
—
Apache-2.0 · source
triton
24.6µs
2.04×
Reported
Apache-2.0 · source
cuda
—
—
Apache-2.0 · source
triton
20.4µs
1.69×
Reported
Apache-2.0 · source
cuda
12.1µs
1.01×
Reproduction-ready
Apache-2.0 · source
triton
61.8µs
5.14×
Reported
Apache-2.0 · source
torch / matmul075b0dFlashInfer-Bench baselines
python
12.0µs
1.00×
Reported
Apache-2.0 · source

Semantics

Inputs and outputs
afp16 [m, k]
bfp16 [n, k]
cfp16 [m, n]
Axes and behavior
kconstant = 2048
mvariable
nconstant = 5120
determinismunspecified
constraintsNo mutation or aliasing
Identity
aliasgemm_n5120_k2048modelqwen3-30b-a3bsha256b1e5273a3389…

Also indexed as 009-gemm-n5120-k2048 · reviewed as the same computation; every definition's cohorts are shown here, ranked separately.

Sources: FlashInfer-Bench (2025-10-20) · Apache-2.0 · NVIDIA SOL-ExecBench (2026-08-21)last observed 2026-08-21How records are decidedJSON