Skip to content
KernelIndex
Search⌘K

Adaptive instance normalization 2d backward

35 eligible runs
rmsnorm

Backward pass for Adaptive Instance Normalization 2D. Computes gradients with respect to input, weight (gamma), and bias (beta). The normalization is performed per-channel across spatial dimensions: y = gamma * (x - mean) / sqrt(var + eps) + beta.

Fastest reported
102.9µsmean · 91% of SOL
RIAC as wellLicense unknown · unknown

Reported evidence · last observed 2026-08-20. 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 16 cases · mean latency · Protocol SOL-ExecBench evaluation stack v1.1 · mean · 21 results · last observed 2026-08-24Record history →
#
Implementation
Latency
vs #1
Trust
Observed
1
102.9µs91% SOL
1.00×
Reported · license unknown · no source
2026-08-20

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

no sourcelicense unknownno install recipeRun detail →
2
138.9µs87% SOL
1.35×
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 →
3
206.8µs81% SOL
2.01×
Reported · license unknown · no source
2026-08-24

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

no sourcelicense unknownno install recipeRun detail →
4
208.4µs81% SOL
2.02×
Reported · license unknown · no source
2026-08-23

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

no sourcelicense unknownno install recipeRun detail →
5
209.4µs81% SOL
2.03×
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 →
6
213.9µs80% SOL
2.08×
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 →
7
220.9µs80% SOL
2.15×
Reported · license unknown · no source
2026-08-17

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

no sourcelicense unknownno install recipeRun detail →
8
222.2µs79% SOL
2.16×
Reported · license unknown · no source
2026-08-18

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

no sourcelicense unknownno install recipeRun detail →
9
224.0µs79% SOL
2.18×
Reported · license unknown · no source
2026-08-08

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

no sourcelicense unknownno install recipeRun detail →
10
225.6µs79% SOL
2.19×
Reported · license unknown · no source
2026-08-12

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

no sourcelicense unknownno install recipeRun detail →
11
232.5µs79% SOL
2.26×
Reported · license unknown · no source
2026-08-16

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

no sourcelicense unknownno install recipeRun detail →
12
257.9µs77% SOL
2.51×
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
258.0µs77% SOL
2.51×
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 →
14
262.1µs77% SOL
2.55×
Reported · license unknown · no source
2026-07-19

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

no sourcelicense unknownno install recipeRun detail →
15
266.8µs76% SOL
2.59×
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 →
16
275.5µs76% SOL
2.68×
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 →
17
313.5µs73% SOL
3.05×
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 →
18
325.1µs72% SOL
3.16×
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 →
19
350.2µs71% SOL
3.40×
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 →
20
397.0µs68% SOL
3.86×
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 →
21
402.4µs68% SOL
3.91×
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 →

Implementations

Implementation
Runtime
Best latency
Evidence
Availability
unknown
102.9µs
1.00×
Reported
license unknown · no source
unknown
206.8µs
2.01×
Reported
license unknown · no source
unknown
232.5µs
2.26×
Reported
license unknown · no source
unknown
256.6µs
2.49×
Reported
license unknown · no source
unknown
262.1µs
2.55×
Reported
license unknown · no source
unknown
266.8µs
2.59×
Reported
license unknown · no source
unknown
275.5µs
2.68×
Reported
license unknown · no source
unknown
307.9µs
2.99×
Reported
license unknown · no source
Show all 35 implementations ›
unknown
550.5µs
5.35×
Reported
license unknown · no source
unknown
623.2µs
6.05×
Reported
license unknown · no source
unknown
1.28ms
12.47×
Reported
license unknown · no source
unknown
208.4µs
2.02×
Reported
license unknown · no source
unknown
209.8µs
2.04×
Reported
license unknown · no source
unknown
214.3µs
2.08×
Reported
license unknown · no source
unknown
220.9µs
2.15×
Reported
license unknown · no source
unknown
222.2µs
2.16×
Reported
license unknown · no source
unknown
224.0µs
2.18×
Reported
license unknown · no source
unknown
225.6µs
2.19×
Reported
license unknown · no source
unknown
232.5µs
2.26×
Reported
license unknown · no source
unknown
234.3µs
2.28×
Reported
license unknown · no source
unknown
258.0µs
2.51×
Reported
license unknown · no source
unknown
257.9µs
2.51×
Reported
license unknown · no source
unknown
313.5µs
3.05×
Reported
license unknown · no source
unknown
322.8µs
3.14×
Reported
license unknown · no source
unknown
325.1µs
3.16×
Reported
license unknown · no source
unknown
338.2µs
3.29×
Reported
license unknown · no source
unknown
350.2µs
3.40×
Reported
license unknown · no source
unknown
378.7µs
3.68×
Reported
license unknown · no source
unknown
397.0µs
3.86×
Reported
license unknown · no source
unknown
402.4µs
3.91×
Reported
license unknown · no source
unknown
680.4µs
6.61×
Reported
license unknown · no source
unknown
948.2µs
9.21×
Reported
license unknown · no source
unknown
138.9µs
1.35×
Reported
license unknown · no source
unknown
209.4µs
2.03×
Reported
license unknown · no source
unknown
213.9µs
2.08×
Reported
license unknown · no source

Semantics

Inputs and outputs
grad_outputfp32 [n, c, h, w]
xfp32 [n, c, h, w]
weightfp32 [c]
meanfp32 [n, c, 1, 1]
stdfp32 [n, c, 1, 1]
grad_inputfp32 [n, c, h, w]
grad_weightfp32 [c]
grad_biasfp32 [c]
Axes and behavior
cconstant = 32
hvariable
nvariable
wvariable
determinismunspecified
constraintsNo mutation or aliasing
Identity
alias039_adaptive_instance_normalization_2d_backwardmodelnafnetsha2563cc491fd1eb9…
Sources: NVIDIA SOL-ExecBench (2026-08-24)last observed 2026-08-24How records are decidedJSON