Conjectures.io

Entry #149

Completed
Hotkey
5HRBtg…qogaLf
Sent
3 Nov 2026, 09:33 UTC
Source
ad0a6e41c852

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 #91Done2 of 2 runs0.809×QualifiesGain lower bound 18.6%Guard passed
MLPvs #92Done2 of 2 runs0.913×QualifiesGain lower bound 6.7%Guard passed
ResNetvs #89Done2 of 2 runs0.571×QualifiesGain lower bound 41.9%Guard passed
MobileNetvs #90Done2 of 2 runs0.639×QualifiesGain lower bound 33.6%Guard passed
Normalized MLPvs #93Done2 of 2 runs0.866×Guard failedGain lower bound 12.0%Guard failed
Mixervs #94Done2 of 2 runs0.857×Guard failedGain lower bound 13.0%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 24330294920.828×Done1 attempt
  2. prod-arm64Seed 24330294920.790×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.890×
  • Batch 1, sparse0.885×
  • Batch 1, wide0.885×
  • Batch 8, normal0.950×
  • Batch 8, sparse0.955×
  • Batch 8, wide0.958×

Mixer

  • Batch 1, normal0.800×
  • Batch 1, sparse0.791×
  • Batch 1, wide0.800×
  • Batch 4, normal0.878×
  • Batch 4, sparse0.886×
  • Batch 4, wide0.879×

ResNet

  • Batch 1, normal0.783×
  • Batch 1, sparse0.789×
  • Batch 1, wide0.785×
  • Batch 4, normal0.835×
  • Batch 4, sparse0.831×
  • Batch 4, wide0.823×

MobileNet

  • Batch 1, normal0.624×
  • Batch 1, sparse0.617×
  • Batch 1, wide0.611×
  • Batch 4, normal0.616×
  • Batch 4, sparse0.637×
  • Batch 4, wide0.609×

Normalized MLP

  • Batch 1, normal0.808×
  • Batch 1, sparse0.810×
  • Batch 1, wide0.805×
  • Batch 8, normal0.956×
  • Batch 8, sparse0.950×
  • Batch 8, wide0.956×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 5676553490.914×Done1 attempt
  2. prod-arm64Seed 5676553490.913×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.876×
  • Batch 1, sparse0.879×
  • Batch 1, wide0.880×
  • Batch 8, normal0.962×
  • Batch 8, sparse0.944×
  • Batch 8, wide0.948×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 36206703120.557×Done1 attempt
  2. prod-arm64Seed 36206703120.586×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.849×
  • Batch 1, sparse0.655×
  • Batch 1, wide0.363×
  • Batch 4, normal0.741×
  • Batch 4, sparse0.646×
  • Batch 4, wide0.376×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 42199372160.614×Done1 attempt
  2. prod-arm64Seed 42199372160.664×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.812×
  • Batch 1, sparse0.639×
  • Batch 1, wide0.487×
  • Batch 4, normal0.786×
  • Batch 4, sparse0.696×
  • Batch 4, wide0.515×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 33947650350.884×Done1 attempt
  2. prod-arm64Seed 33947650350.848×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal0.798×
  • Batch 1, sparse0.796×
  • Batch 1, wide0.796×
  • Batch 8, normal0.946×
  • Batch 8, sparse0.957×
  • Batch 8, wide0.934×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 6181620230.880×Done1 attempt
  2. prod-arm64Seed 6181620230.834×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.779×
  • Batch 1, sparse0.782×
  • Batch 1, wide0.783×
  • Batch 4, normal0.898×
  • Batch 4, sparse0.969×
  • Batch 4, wide0.899×