Skip to content
KernelIndex
Search⌘K

Video latent GELU activation

34 eligible runs
mlp

GELU activation function using tanh approximation applied element-wise to video latent tensors. Formula: 0.5 * x * (1 + tanh(sqrt(2/pi) * (x + 0.044715 * x^3))). This is executed in transformer feedforward networks on intermediate tensors.

Fastest reported
19.7µsmean · 65% of SOL
RIAC as wellLicense unknown · unknown

Reported evidence · last observed 2026-08-15. Reported by source; not independently reproduced. No public source. License unknown.

No public sourceRun detail →

Current records

Workloadsuite of 16 cases · mean latencyintermediate_size = 409616 cases

Not measured on H100 for this workload. Challenges →

Source-native comparison · GPU NVIDIA B200 · Workload suite of 16 cases · mean latency · Protocol SOL-ExecBench evaluation stack v1.1 · mean · 20 results · last observed 2026-08-15Record history →
#
Implementation
Latency
vs #1
Trust
Observed
1
19.7µs65% SOL
1.00×
Reported · license unknown · no source
2026-08-15

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

no sourcelicense unknownno install recipeRun detail →
2
19.9µs64% SOL
1.01×
Reported · license unknown · no source
2026-08-09

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

no sourcelicense unknownno install recipeRun detail →
3
20.1µs63% SOL
1.02×
Reported · license unknown · no source
2026-08-07

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

no sourcelicense unknownno install recipeRun detail →
4
20.1µs62% SOL
1.02×
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 →
5
20.2µs62% SOL
1.02×
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 →
6
20.2µs61% SOL
1.03×
Reported · license unknown · no source
2026-08-03

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

no sourcelicense unknownno install recipeRun detail →
7
20.4µs61% SOL
1.03×
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 →
8
20.5µs60% SOL
1.04×
Reported · license unknown · no source
2026-08-03

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

no sourcelicense unknownno install recipeRun detail →
9
20.9µs60% 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 →
10
21.0µs59% SOL
1.07×
Reported · license unknown · no source
2026-07-15

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

no sourcelicense unknownno install recipeRun detail →
11
21.1µs59% SOL
1.07×
Reported · license unknown · no source
2026-07-17

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

no sourcelicense unknownno install recipeRun detail →
12
21.3µs58% 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 →
13
21.4µs56% SOL
1.09×
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 →
14
23.1µs49% SOL
1.17×
Reported · license unknown · no source
2026-07-17

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

no sourcelicense unknownno install recipeRun detail →
15
23.2µs48% SOL
1.18×
Reported · license unknown · no source
2026-07-17

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

no sourcelicense unknownno install recipeRun detail →
16
23.2µs49% SOL
1.18×
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 →
17
23.4µs48% SOL
1.19×
Reported · license unknown · no source
2026-07-20

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

no sourcelicense unknownno install recipeRun detail →
18
24.4µs45% SOL
1.24×
Reported · license unknown · no source
2026-08-14

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

no sourcelicense unknownno install recipeRun detail →
19
24.8µs44% SOL
1.26×
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 →
20
25.1µs42% SOL
1.28×
Reported · license unknown · no source
2026-08-13

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
1.00×
Reported
license unknown · no source
unknown
20.2µs
1.02×
Reported
license unknown · no source
unknown
20.2µs
1.03×
Reported
license unknown · no source
unknown
20.3µs
1.03×
Reported
license unknown · no source
unknown
20.9µs
1.06×
Reported
license unknown · no source
unknown
21.0µs
1.07×
Reported
license unknown · no source
unknown
21.9µs
1.11×
Reported
license unknown · no source
unknown
22.2µs
1.13×
Reported
license unknown · no source
Show all 34 implementations ›
unknown
22.8µs
1.16×
Reported
license unknown · no source
unknown
22.8µs
1.16×
Reported
license unknown · no source
unknown
23.1µs
1.17×
Reported
license unknown · no source
unknown
23.4µs
1.19×
Reported
license unknown · no source
unknown
24.4µs
1.24×
Reported
license unknown · no source
unknown
25.8µs
1.31×
Reported
license unknown · no source
unknown
24.8µs
1.26×
Reported
license unknown · no source
unknown
236.9µs
12.03×
Reported
license unknown · no source
unknown
20.2µs
1.03×
Reported
license unknown · no source
unknown
20.2µs
1.03×
Reported
license unknown · no source
unknown
20.4µs
1.03×
Reported
license unknown · no source
unknown
20.5µs
1.04×
Reported
license unknown · no source
unknown
20.9µs
1.06×
Reported
license unknown · no source
unknown
21.1µs
1.07×
Reported
license unknown · no source
unknown
21.2µs
1.08×
Reported
license unknown · no source
unknown
21.3µs
1.08×
Reported
license unknown · no source
unknown
21.4µs
1.09×
Reported
license unknown · no source
unknown
21.6µs
1.10×
Reported
license unknown · no source
unknown
22.8µs
1.16×
Reported
license unknown · no source
unknown
22.9µs
1.16×
Reported
license unknown · no source
unknown
23.2µs
1.18×
Reported
license unknown · no source
unknown
23.2µs
1.18×
Reported
license unknown · no source
unknown
25.1µs
1.28×
Reported
license unknown · no source
unknown
19.9µs
1.01×
Reported
license unknown · no source
unknown
20.1µs
1.02×
Reported
license unknown · no source
unknown
20.1µs
1.02×
Reported
license unknown · no source

Semantics

Inputs and outputs
xfp32 [batch_size, seq_len, intermediate_size]
outputfp32 [batch_size, seq_len, intermediate_size]
Axes and behavior
seq_lenvariable
batch_sizevariable
intermediate_sizeconstant = 4096
determinismunspecified
constraintsNo mutation or aliasing
Identity
alias025_video_latent_gelu_activationmodelwan2-2-t2v-a14bsha25645c92c88fa46…
Sources: NVIDIA SOL-ExecBench (2026-08-15)last observed 2026-08-15How records are decidedJSON