Entry #98
Completed- Hotkey
- 5HEoCg…GF71rP
- Sent
- 12 Nov 2026, 09:45 UTC
- Source
- d7248d99c53d
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 | 17.006×Generated 17.517×, real models 16.509× | Above the barGain lower bound -3074.7%Guard passed |
| MLPvs #44 | Done2 of 2 runs | 0.983×Generated 1.024×, real models 0.943× | Above the barGain lower bound -25.7%Guard passed |
| ResNetvs #51 | Done2 of 2 runs | 4.017×Generated 4.563×, real models 3.536× | Above the barGain lower bound -364.2%Guard passed |
| MobileNetvs #53 | Done2 of 2 runs | 1.011×Generated 1.088×, real models 0.939× | Above the barGain lower bound -30.6%Guard passed |
| Normalized MLPvs #48 | Done2 of 2 runs | 40.475×Generated 58.990×, real models 27.770× | Above the barGain lower bound -5025.1%Guard passed |
| Mixervs #47 | Done2 of 2 runs | 160.159×Generated 176.146×, real models 145.623× | Above the barGain lower bound -19263.0%Guard passed |
| Recurrentvs #55 | Done2 of 2 runs | 823.487×Generated 715.175×, real models 948.204× | Above the barGain lower bound -88325.8%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 17.517× · real models 16.509×
MachineTime
- prod-amd64Seed 96835536516.757×Done1 attempt0.171× of the baseline kernels' time
- prod-arm64Seed 96835536517.258×Done1 attempt0.185× of the baseline kernels' time
By model and inputGuard at 1.100×
MLP
- Batch 1, normal1.302×
- Batch 1, sparse0.881×
- Batch 1, wide1.470×
Mixer
- Batch 1, normal165.118×
- Batch 1, sparse167.507×
- Batch 1, wide194.159×
ResNet
- Batch 1, normal3.838×
- Batch 1, sparse4.297×
- Batch 1, wide3.857×
MobileNet
- Batch 1, normal0.966×
- Batch 1, sparse0.958×
- Batch 1, wide1.003×
Recurrent
- Batch 1, normal721.801×
- Batch 1, sparse660.177×
- Batch 1, wide675.135×
Normalized MLP
- Batch 1, normal52.189×
- Batch 1, sparse52.703×
- Batch 1, wide54.435×
Real models
- appliance-energy-mixer on Appliance Energy Prediction250.887×
- compact-mixer on digits150.197×
- compact-mixer on fashion167.073×
- compact-mixer on mnist158.309×
- compact-mlp on digits1.105×
- compact-mlp on fashion2.544×
- compact-mlp on mnist0.786×
- compact-mobilenet on digits0.926×
- compact-mobilenet on fashion0.829×
- compact-mobilenet on mnist0.793×
- compact-normalized_mlp on digits77.226×
- compact-normalized_mlp on fashion23.859×
- compact-normalized_mlp on mnist36.512×
- compact-resnet on digits2.779×
- compact-resnet on fashion3.515×
- compact-resnet on mnist3.522×
- appliance-gru on appliance-energy750.134×
- mitbih-gru on mitbih897.008×
- mnist-rows-lstm on mnist-rows709.938×
- speech-commands-gru on speech-commands694.008×
- speech-commands-lstm-gru on speech-commands854.417×
- spoken-arabic-digit-lstm on spoken-arabic-digit1101.911×
- tiny-shakespeare-lstm on tiny-shakespeare832.312×
- uci-har-gru on uci-har828.909×
- uci-har-lstm on uci-har1151.055×
- uci-har-mixer on UCI-HAR170.053×
- uci-har-mlp on UCI-HAR1.066×
- uci-har-mobilenet on UCI-HAR0.927×
- uci-har-normalized_mlp on UCI-HAR22.789×
- uci-har-resnet on UCI-HAR4.635×
MLP2 runs
RunsGenerated 1.024× · real models 0.943×
MachineTime
- prod-amd64Seed 9683553650.803×Done1 attempt0.015× of the baseline kernels' time
- prod-arm64Seed 9683553651.203×Done1 attempt0.022× of the baseline kernels' time
By model and inputGuard at 1.100×
MLP
- Batch 1, normal0.974×
- Batch 1, sparse1.185×
- Batch 1, wide1.095×
Real models
- compact-mlp on digits0.626×
- compact-mlp on fashion0.832×
- compact-mlp on mnist1.667×
- uci-har-mlp on UCI-HAR0.912×
ResNet2 runs
RunsGenerated 4.563× · real models 3.536×
MachineTime
- prod-amd64Seed 9683553654.094×Done1 attempt0.035× of the baseline kernels' time
- prod-arm64Seed 9683553653.941×Done1 attempt0.034× of the baseline kernels' time
By model and inputGuard at 1.100×
ResNet
- Batch 1, normal5.247×
- Batch 1, sparse4.947×
- Batch 1, wide4.046×
Real models
- compact-resnet on digits2.803×
- compact-resnet on fashion3.564×
- compact-resnet on mnist3.507×
- uci-har-resnet on UCI-HAR4.461×
MobileNet2 runs
RunsGenerated 1.088× · real models 0.939×
MachineTime
- prod-amd64Seed 9683553650.915×Done1 attempt0.011× of the baseline kernels' time
- prod-arm64Seed 9683553651.117×Done1 attempt0.010× of the baseline kernels' time
By model and inputGuard at 1.100×
MobileNet
- Batch 1, normal0.973×
- Batch 1, sparse0.955×
- Batch 1, wide1.305×
Real models
- compact-mobilenet on digits0.929×
- compact-mobilenet on fashion0.956×
- compact-mobilenet on mnist0.956×
- uci-har-mobilenet on UCI-HAR0.917×
Normalized MLP2 runs
RunsGenerated 58.990× · real models 27.770×
MachineTime
- prod-amd64Seed 96835536548.733×Done1 attempt0.524× of the baseline kernels' time
- prod-arm64Seed 96835536533.616×Done1 attempt0.514× of the baseline kernels' time
By model and inputGuard at 1.100×
Normalized MLP
- Batch 1, normal58.335×
- Batch 1, sparse57.301×
- Batch 1, wide59.614×
Real models
- compact-normalized_mlp on digits44.570×
- compact-normalized_mlp on fashion18.419×
- compact-normalized_mlp on mnist33.016×
- uci-har-normalized_mlp on UCI-HAR21.943×
Mixer2 runs
RunsGenerated 176.146× · real models 145.623×
MachineTime
- prod-amd64Seed 968355365172.232×Done1 attempt0.871× of the baseline kernels' time
- prod-arm64Seed 968355365148.932×Done1 attempt0.842× of the baseline kernels' time
By model and inputGuard at 1.100×
Mixer
- Batch 1, normal176.384×
- Batch 1, sparse175.818×
- Batch 1, wide197.490×
Real models
- appliance-energy-mixer on Appliance Energy Prediction223.048×
- compact-mixer on digits84.376×
- compact-mixer on fashion155.343×
- compact-mixer on mnist153.241×
- uci-har-mixer on UCI-HAR146.172×
Recurrent2 runs
RunsGenerated 715.175× · real models 948.204×
MachineTime
- prod-amd64Seed 968355365853.812×Done1 attempt8.323× of the baseline kernels' time
- prod-arm64Seed 968355365794.239×Done1 attempt8.493× of the baseline kernels' time
By model and inputGuard at 1.100×
Recurrent
- Batch 1, normal770.677×
- Batch 1, sparse702.177×
- Batch 1, wide709.494×
Real models
- appliance-gru on appliance-energy741.532×
- mitbih-gru on mitbih972.081×
- mnist-rows-lstm on mnist-rows787.882×
- speech-commands-gru on speech-commands770.019×
- speech-commands-lstm-gru on speech-commands967.495×
- spoken-arabic-digit-lstm on spoken-arabic-digit1298.723×
- tiny-shakespeare-lstm on tiny-shakespeare918.892×
- uci-har-gru on uci-har919.142×
- uci-har-lstm on uci-har1335.082×