Entry #87
Completed- Hotkey
- 5EP4gp…PRs6tS
- Sent
- 12 Nov 2026, 06:11 UTC
- Source
- da8e02a0f461
Progress
- Queued
- Proof
- Planned
- Timing
- Done
Done means every board has a decision. It does not mean the entry took a board over.
On each board
| Board | Runs | Time | Decision |
|---|---|---|---|
| Overallvs #50 | Done2 of 2 runs | 20.416×Generated 21.953×, real models 18.986× | Above the barGain lower bound -2140.3%Guard passed |
| MLPvs #44 | Done2 of 2 runs | 1.030×Generated 0.905×, real models 1.172× | Above the barGain lower bound -31.3%Guard passed |
| ResNetvs #51 | Done2 of 2 runs | 111.364×Generated 126.575×, real models 97.981× | Above the barGain lower bound -11803.1%Guard passed |
| MobileNetvs #53 | Done2 of 2 runs | 89.461×Generated 96.206×, real models 83.188× | Above the barGain lower bound -10350.9%Guard passed |
| Normalized MLPvs #48 | Done2 of 2 runs | 38.903×Generated 48.449×, real models 31.237× | Above the barGain lower bound -4710.3%Guard passed |
| Mixervs #47 | Done2 of 2 runs | 167.687×Generated 188.013×, real models 149.557× | Above the barGain lower bound -18130.2%Guard passed |
| Recurrentvs #54 | Done2 of 2 runs | 0.982×Generated 0.974×, real models 0.990× | Above the barGain lower bound -7.8%Guard passed |
The source stays private while it holds a board, and for entries that never held one.
Every run
Each board times the entry on every benchmark machine, over fresh seeds, on generated workloads and on trained models with real data. A run's time combines the two suites by their published weights; the bars below split it by model, batch and input, and by real model.
Overall2 runs
RunsGenerated 21.953× · real models 18.986×
MachineTime
- prod-amd64Seed 70323644520.914×Done1 attempt0.207× of the baseline kernels' time
- prod-arm64Seed 70323644519.930×Done1 attempt0.208× of the baseline kernels' time
By model and inputGuard at 1.100×
MLP
- Batch 1, normal1.112×
- Batch 1, sparse0.966×
- Batch 1, wide0.972×
Mixer
- Batch 1, normal173.784×
- Batch 1, sparse167.757×
- Batch 1, wide208.361×
ResNet
- Batch 1, normal115.704×
- Batch 1, sparse134.107×
- Batch 1, wide115.607×
MobileNet
- Batch 1, normal93.678×
- Batch 1, sparse94.493×
- Batch 1, wide93.424×
Recurrent
- Batch 1, normal0.938×
- Batch 1, sparse0.922×
- Batch 1, wide0.906×
Normalized MLP
- Batch 1, normal53.290×
- Batch 1, sparse53.581×
- Batch 1, wide65.091×
Real models
- appliance-energy-mixer on Appliance Energy Prediction262.745×
- compact-mixer on digits161.426×
- compact-mixer on fashion153.759×
- compact-mixer on mnist155.190×
- compact-mlp on digits0.914×
- compact-mlp on fashion2.226×
- compact-mlp on mnist1.095×
- compact-mobilenet on digits59.003×
- compact-mobilenet on fashion83.326×
- compact-mobilenet on mnist76.155×
- compact-normalized_mlp on digits63.530×
- compact-normalized_mlp on fashion30.515×
- compact-normalized_mlp on mnist34.411×
- compact-resnet on digits76.700×
- compact-resnet on fashion121.095×
- compact-resnet on mnist118.730×
- appliance-gru on appliance-energy0.943×
- mitbih-gru on mitbih0.948×
- mnist-rows-lstm on mnist-rows0.711×
- speech-commands-gru on speech-commands0.943×
- speech-commands-lstm-gru on speech-commands0.954×
- spoken-arabic-digit-lstm on spoken-arabic-digit0.927×
- tiny-shakespeare-lstm on tiny-shakespeare0.770×
- uci-har-gru on uci-har0.954×
- uci-har-lstm on uci-har0.918×
- uci-har-mixer on UCI-HAR167.407×
- uci-har-mlp on UCI-HAR1.028×
- uci-har-mobilenet on UCI-HAR73.524×
- uci-har-normalized_mlp on UCI-HAR21.513×
- uci-har-resnet on UCI-HAR79.203×
MLP2 runs
RunsGenerated 0.905× · real models 1.172×
MachineTime
- prod-amd64Seed 7032364450.911×Done1 attempt0.016× of the baseline kernels' time
- prod-arm64Seed 7032364451.164×Done1 attempt0.013× of the baseline kernels' time
By model and inputGuard at 1.100×
MLP
- Batch 1, normal0.965×
- Batch 1, sparse0.965×
- Batch 1, wide0.943×
Real models
- compact-mlp on digits1.094×
- compact-mlp on fashion1.668×
- compact-mlp on mnist1.075×
- uci-har-mlp on UCI-HAR0.962×
ResNet2 runs
RunsGenerated 126.575× · real models 97.981×
MachineTime
- prod-amd64Seed 703236445118.052×Done1 attempt0.994× of the baseline kernels' time
- prod-arm64Seed 703236445105.055×Done1 attempt0.999× of the baseline kernels' time
By model and inputGuard at 1.100×
ResNet
- Batch 1, normal116.269×
- Batch 1, sparse139.145×
- Batch 1, wide120.412×
Real models
- compact-resnet on digits76.692×
- compact-resnet on fashion121.384×
- compact-resnet on mnist119.673×
- uci-har-resnet on UCI-HAR82.730×
MobileNet2 runs
RunsGenerated 96.206× · real models 83.188×
MachineTime
- prod-amd64Seed 70323644592.325×Done1 attempt1.199× of the baseline kernels' time
- prod-arm64Seed 70323644586.685×Done1 attempt1.155× of the baseline kernels' time
By model and inputGuard at 1.100×
MobileNet
- Batch 1, normal97.199×
- Batch 1, sparse96.675×
- Batch 1, wide93.575×
Real models
- compact-mobilenet on digits64.968×
- compact-mobilenet on fashion94.929×
- compact-mobilenet on mnist93.408×
- uci-har-mobilenet on UCI-HAR83.132×
Normalized MLP2 runs
RunsGenerated 48.449× · real models 31.237×
MachineTime
- prod-amd64Seed 70323644543.041×Done1 attempt0.503× of the baseline kernels' time
- prod-arm64Seed 70323644535.162×Done1 attempt0.491× of the baseline kernels' time
By model and inputGuard at 1.100×
Normalized MLP
- Batch 1, normal48.860×
- Batch 1, sparse47.139×
- Batch 1, wide52.485×
Real models
- compact-normalized_mlp on digits74.369×
- compact-normalized_mlp on fashion22.412×
- compact-normalized_mlp on mnist26.650×
- uci-har-normalized_mlp on UCI-HAR21.435×
Mixer2 runs
RunsGenerated 188.013× · real models 149.557×
MachineTime
- prod-amd64Seed 703236445165.831×Done1 attempt0.876× of the baseline kernels' time
- prod-arm64Seed 703236445169.563×Done1 attempt0.859× of the baseline kernels' time
By model and inputGuard at 1.100×
Mixer
- Batch 1, normal176.937×
- Batch 1, sparse184.074×
- Batch 1, wide215.629×
Real models
- appliance-energy-mixer on Appliance Energy Prediction232.041×
- compact-mixer on digits89.552×
- compact-mixer on fashion155.879×
- compact-mixer on mnist157.920×
- uci-har-mixer on UCI-HAR146.275×
Recurrent2 runs
RunsGenerated 0.974× · real models 0.990×
MachineTime
- prod-amd64Seed 7032364451.014×Done1 attempt0.011× of the baseline kernels' time
- prod-arm64Seed 7032364450.951×Done1 attempt0.012× of the baseline kernels' time
By model and inputGuard at 1.100×
Recurrent
- Batch 1, normal0.939×
- Batch 1, sparse0.983×
- Batch 1, wide0.971×
Real models
- appliance-gru on appliance-energy0.924×
- mitbih-gru on mitbih0.953×
- mnist-rows-lstm on mnist-rows1.045×
- speech-commands-gru on speech-commands0.929×
- speech-commands-lstm-gru on speech-commands0.962×
- spoken-arabic-digit-lstm on spoken-arabic-digit1.024×
- tiny-shakespeare-lstm on tiny-shakespeare1.124×
- uci-har-gru on uci-har0.937×
- uci-har-lstm on uci-har1.033×