Skip to content
KernelIndex
Search⌘K
Sign in

ResNet block with time embedding

23 eligible runs
diffusion

ResNet block with time embedding injection for Stable Diffusion UNet. Performs: GroupNorm -> SiLU -> Conv2d -> time_emb projection -> addition -> GroupNorm -> SiLU -> Dropout -> Conv2d -> residual addition. This is a fundamental building block that appears in every down/up block of Stable Diffusion.

Fastest reported
85.5µsmean · 73% of SOL
RIAC as wellLicense unknown · unknown

Reported evidence · last observed 2026-08-18. Reported by source; not independently reproduced. No public source. License unknown.

No public sourceRun detail →

Current records

Workloadsuite of 16 cases · mean latencyin_channels = 320 · out_channels = 320 · norm_num_groups = 32 · time_emb_channels = 128016 cases
height
width
batch_size
dtype
16161fp32161664fp3232322fp3232324fp3237378fp3241414fp3248481fp3248484fp3253531fp3264642fp3264644fp3267672fp3296962fp3296964fp321281281fp321311311fp32
scrolls · 16 cases total
Aggregatessuite of 16 cases · mean latency
HardwareBest knownImplementationRuns
NVIDIA B200 · env 185.5 µsRIAC as well16NVIDIA B200 · env 2134.1 µsRecursive7

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 · 16 results · last observed 2026-08-24Record history →
#
Implementation
Latency
vs #1
Trust
Observed
1
RIAC as well
85.5µs73% SOL
1.00×
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 →
2
Amir M. Mir | SF Tensor
98.3µs69% SOL
1.15×
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 →
3
ac4k
104.0µs68% SOL
1.22×
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 →
4
Hyra
128.3µs62% SOL
1.50×
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 →
5
doubleAI
133.7µs61% SOL
1.56×
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
Proud Wren @ NVIDIA
147.3µs59% SOL
1.72×
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
Geometric
148.2µs58% SOL
1.73×
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 →
8
Geometric
160.1µs56% SOL
1.87×
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 →
9
Amir M. Mir | SF Tensor
181.7µs53% SOL
2.12×
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 →
10
Recursive
190.0µs52% SOL
2.22×
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 →
11
Cursor
209.0µs49% SOL
2.44×
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 →
12
Geometric
228.9µs46% SOL
2.68×
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 →
13
Geometric
238.9µs45% SOL
2.79×
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 →
14
Geometric
459.5µs29% SOL
5.37×
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 →
15
Geometric
474.6µs29% SOL
5.55×
Reported · license unknown · no source
2026-08-11

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

no sourcelicense unknownno install recipeRun detail →
16
Geometric
244.1ms0% SOL
2852.94×
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 →

Implementations

Implementation
Runtime
Best latency
Evidence
Availability
RIAC as well
unknown
85.5µs
1.00×
Reported
license unknown · no source
Recursive
unknown
134.1µs
1.57×
Reported
license unknown · no source
doubleAI
unknown
133.7µs
1.56×
Reported
license unknown · no source
Proud Wren @ NVIDIA
unknown
142.2µs
1.66×
Reported
license unknown · no source
Geometric
unknown
148.2µs
1.73×
Reported
license unknown · no source
Cursor
unknown
193.4µs
2.26×
Reported
license unknown · no source
doubleAI
unknown
137.9µs
1.61×
Reported
license unknown · no source
Proud Wren @ NVIDIA
unknown
147.3µs
1.72×
Reported
license unknown · no source
Show all 23 implementations ›
doubleAI
unknown
147.4µs
1.72×
Reported
license unknown · no source
Geometric
unknown
160.1µs
1.87×
Reported
license unknown · no source
Amir M. Mir | SF Tensor
unknown
176.0µs
2.06×
Reported
license unknown · no source
Amir M. Mir | SF Tensor
unknown
181.7µs
2.12×
Reported
license unknown · no source
Recursive
unknown
190.0µs
2.22×
Reported
license unknown · no source
Amir M. Mir | SF Tensor
unknown
191.1µs
2.23×
Reported
license unknown · no source
Cursor
unknown
209.0µs
2.44×
Reported
license unknown · no source
Geometric
unknown
228.9µs
2.68×
Reported
license unknown · no source
Geometric
unknown
238.9µs
2.79×
Reported
license unknown · no source
Geometric
unknown
459.5µs
5.37×
Reported
license unknown · no source
Geometric
unknown
474.6µs
5.55×
Reported
license unknown · no source
Geometric
unknown
244.1ms
2852.94×
Reported
license unknown · no source
Amir M. Mir | SF Tensor
unknown
98.3µs
1.15×
Reported
license unknown · no source
ac4k
unknown
104.0µs
1.22×
Reported
license unknown · no source
Hyra
unknown
128.3µs
1.50×
Reported
license unknown · no source

Semantics

Inputs and outputs
xfp32 [batch_size, in_channels, height, width]
time_embfp32 [batch_size, time_emb_channels]
norm1_weightfp32 [in_channels]
norm1_biasfp32 [in_channels]
conv1_weightfp32 [out_channels, in_channels, 3, 3]
conv1_biasfp32 [out_channels]
time_emb_proj_weightfp32 [out_channels, time_emb_channels]
time_emb_proj_biasfp32 [out_channels]
norm2_weightfp32 [out_channels]
norm2_biasfp32 [out_channels]
conv2_weightfp32 [out_channels, out_channels, 3, 3]
conv2_biasfp32 [out_channels]
norm_epsfp32 scalar
outputfp32 [batch_size, out_channels, height, width]
Axes and behavior
widthvariable
heightvariable
batch_sizevariable
in_channelsconstant = 320
out_channelsconstant = 320
norm_num_groupsconstant = 32
time_emb_channelsconstant = 1280
determinismunspecified
constraintsNo mutation or aliasing
Identity
alias066_resnet_block_with_time_embeddingmodelstable-diffusion-v1-5sha2564fabd0b1273a…
Sources: NVIDIA SOL-ExecBench (2026-08-24)last observed 2026-08-24How records are decidedJSON