Conjectures.io

Entry #155

Completed
Hotkey
5HQP4G…iLbLKS
Sent
3 Nov 2026, 23:19 UTC
Source
41bcdaae8a58

Progress

  1. Queued
  2. Proof
  3. Planned
  4. Timing
  5. Done

Done means every board has a decision. It does not mean the entry took a board over.

On each board

BoardRunsTimeDecision
Overallvs #108Done2 of 2 runs0.919×Guard failedGain lower bound 7.9%Guard failed
MLPvs #113Done2 of 2 runs1.000×Above the barGain lower bound -0.1%Guard passed
ResNetvs #109Done2 of 2 runs0.849×QualifiesGain lower bound 14.0%Guard passed
MobileNetvs #110Done2 of 2 runs0.903×Guard failedGain lower bound 7.9%Guard failed
Normalized MLPvs #114Done2 of 2 runs0.998×Above the barGain lower bound -0.1%Guard passed
Mixervs #112Done2 of 2 runs1.083×Guard failedGain lower bound -10.5%Guard failed

Download Kernel.lean

Every run

Each board times the entry on every benchmark machine, over fresh seeds. A run's time is its weighted geometric mean over the workloads; the bars below split it by model, batch and input.

Overall2 runs

Runs

MachineTime
  1. prod-amd64Seed 25265973240.921×Done1 attempt
  2. prod-arm64Seed 25265973240.917×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.875×
  • Batch 1, sparse0.929×
  • Batch 1, wide0.891×
  • Batch 8, normal0.838×
  • Batch 8, sparse0.840×
  • Batch 8, wide0.837×

Mixer

  • Batch 1, normal1.008×
  • Batch 1, sparse0.982×
  • Batch 1, wide0.996×
  • Batch 4, normal1.122×
  • Batch 4, sparse1.156×
  • Batch 4, wide1.181×

ResNet

  • Batch 1, normal0.922×
  • Batch 1, sparse0.915×
  • Batch 1, wide0.910×
  • Batch 4, normal0.809×
  • Batch 4, sparse0.823×
  • Batch 4, wide0.824×

MobileNet

  • Batch 1, normal0.975×
  • Batch 1, sparse1.009×
  • Batch 1, wide1.009×
  • Batch 4, normal0.891×
  • Batch 4, sparse0.911×
  • Batch 4, wide0.888×

Normalized MLP

  • Batch 1, normal0.889×
  • Batch 1, sparse0.890×
  • Batch 1, wide0.891×
  • Batch 8, normal0.850×
  • Batch 8, sparse0.851×
  • Batch 8, wide0.856×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 1035319221.002×Done1 attempt
  2. prod-arm64Seed 1035319220.998×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal1.001×
  • Batch 1, sparse1.001×
  • Batch 1, wide1.001×
  • Batch 8, normal0.999×
  • Batch 8, sparse1.000×
  • Batch 8, wide0.999×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 31681106640.860×Done1 attempt
  2. prod-arm64Seed 31681106640.838×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.914×
  • Batch 1, sparse0.895×
  • Batch 1, wide0.895×
  • Batch 4, normal0.806×
  • Batch 4, sparse0.822×
  • Batch 4, wide0.781×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 35737523440.910×Done1 attempt
  2. prod-arm64Seed 35737523440.896×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal1.292×
  • Batch 1, sparse0.903×
  • Batch 1, wide0.885×
  • Batch 4, normal0.884×
  • Batch 4, sparse0.786×
  • Batch 4, wide0.789×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 40615687231.000×Done1 attempt
  2. prod-arm64Seed 40615687230.996×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal0.998×
  • Batch 1, sparse0.997×
  • Batch 1, wide0.997×
  • Batch 8, normal1.000×
  • Batch 8, sparse0.999×
  • Batch 8, wide0.999×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 555105161.108×Done1 attempt
  2. prod-arm64Seed 555105161.058×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.985×
  • Batch 1, sparse0.913×
  • Batch 1, wide1.103×
  • Batch 4, normal1.134×
  • Batch 4, sparse1.316×
  • Batch 4, wide1.120×