Skip to content
KernelIndex
Search⌘K

Post norm residual

31 eligible runs
rmsnorm

Post-normalization residual connection pattern: output = residual + RMSNorm(sublayer_output). RMSNorm applies learned weight scaling after normalization. Used after both attention and MLP blocks in Olmo3 architecture.

Fastest reported
15.5µsmean · 80% of SOL
RecursiveLicense unknown · unknown

Reported evidence · last observed 2026-06-07. Reported by source; not independently reproduced. No public source. License unknown.

No public sourceRun detail →

Current records

Workloadsuite of 16 cases · mean latencyhidden_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 v1.0 · mean · 16 results · last observed 2026-08-18Record history →
#
Implementation
Latency
vs #1
Trust
Observed
1
15.5µs80% SOL
1.00×
Reported · license unknown · no source
2026-06-07

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

no sourcelicense unknownno install recipeRun detail →
2
15.9µs78% SOL
1.02×
Reported · license unknown · no source
2026-07-01

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

no sourcelicense unknownno install recipeRun detail →
3
16.3µs78% SOL
1.05×
Reported · license unknown · no source
2026-07-01

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

no sourcelicense unknownno install recipeRun detail →
4
16.5µs78% SOL
1.06×
Reported · license unknown · no source
2026-06-25

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

no sourcelicense unknownno install recipeRun detail →
5
17.2µs76% SOL
1.11×
Reported · license unknown · no source
2026-07-05

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

no sourcelicense unknownno install recipeRun detail →
6
17.5µs76% SOL
1.13×
Reported · license unknown · no source
2026-05-20

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

no sourcelicense unknownno install recipeRun detail →
7
18.5µs74% SOL
1.19×
Reported · license unknown · no source
2026-04-25

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

no sourcelicense unknownno install recipeRun detail →
8
18.5µs74% SOL
1.19×
Reported · license unknown · no source
2026-07-08

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

no sourcelicense unknownno install recipeRun detail →
9
19.2µs73% SOL
1.24×
Reported · license unknown · no source
2026-04-14

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

no sourcelicense unknownno install recipeRun detail →
10
19.2µs73% SOL
1.24×
Reported · license unknown · no source
2026-07-07

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

no sourcelicense unknownno install recipeRun detail →
11
21.8µs68% SOL
1.40×
Reported · license unknown · no source
2026-05-07

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

no sourcelicense unknownno install recipeRun detail →
12
27.6µs60% SOL
1.78×
Reported · license unknown · no source
2026-04-15

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

no sourcelicense unknownno install recipeRun detail →
13
39.6µs50% SOL
2.55×
Reported · license unknown · no source
2026-05-28

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

no sourcelicense unknownno install recipeRun detail →
14
43.7µs46% SOL
2.82×
Reported · license unknown · no source
2026-04-15

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

no sourcelicense unknownno install recipeRun detail →
15
43.7µs46% SOL
2.82×
Reported · license unknown · no source
2026-04-15

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

no sourcelicense unknownno install recipeRun detail →
16
252.7µs11% SOL
16.28×
Reported · license unknown · no source
2026-04-23

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
15.9µs
1.02×
Reported
license unknown · no source
unknown
15.9µs
1.02×
Reported
license unknown · no source
unknown
15.7µs
1.01×
Reported
license unknown · no source
unknown
16.3µs
1.05×
Reported
license unknown · no source
unknown
16.5µs
1.06×
Reported
license unknown · no source
unknown
16.9µs
1.09×
Reported
license unknown · no source
unknown
17.2µs
1.11×
Reported
license unknown · no source
unknown
17.2µs
1.11×
Reported
license unknown · no source
Show all 31 implementations ›
unknown
17.6µs
1.14×
Reported
license unknown · no source
unknown
18.5µs
1.19×
Reported
license unknown · no source
unknown
19.2µs
1.24×
Reported
license unknown · no source
unknown
19.6µs
1.26×
Reported
license unknown · no source
unknown
20.5µs
1.32×
Reported
license unknown · no source
unknown
21.8µs
1.40×
Reported
license unknown · no source
unknown
27.6µs
1.78×
Reported
license unknown · no source
unknown
39.6µs
2.55×
Reported
license unknown · no source
unknown
16.3µs
1.05×
Reported
license unknown · no source
unknown
16.7µs
1.08×
Reported
license unknown · no source
unknown
17.2µs
1.11×
Reported
license unknown · no source
unknown
17.5µs
1.13×
Reported
license unknown · no source
unknown
18.5µs
1.19×
Reported
license unknown · no source
unknown
19.0µs
1.22×
Reported
license unknown · no source
unknown
18.9µs
1.22×
Reported
license unknown · no source
unknown
19.2µs
1.24×
Reported
license unknown · no source
unknown
19.5µs
1.26×
Reported
license unknown · no source
unknown
19.6µs
1.26×
Reported
license unknown · no source
unknown
19.6µs
1.26×
Reported
license unknown · no source
unknown
43.7µs
2.82×
Reported
license unknown · no source
unknown
43.7µs
2.82×
Reported
license unknown · no source
unknown
252.7µs
16.28×
Reported
license unknown · no source
unknown
15.5µs
1.00×
Reported
license unknown · no source

Semantics

Inputs and outputs
sublayer_outputbf16 [batch_size, seq_len, hidden_size]
residualbf16 [batch_size, seq_len, hidden_size]
weightbf16 [hidden_size]
epsfp32 scalar
outputbf16 [batch_size, seq_len, hidden_size]
Axes and behavior
seq_lenvariable
batch_sizevariable
hidden_sizeconstant = 4096
determinismunspecified
constraintsNo mutation or aliasing
Identity
alias033_post_norm_residualmodelolmo-3-1025-7bsha25609665155af3a…
Sources: NVIDIA SOL-ExecBench (2026-08-18)last observed 2026-08-18How records are decidedJSON