Conjectures.io

Entry #147

Completed
Hotkey
5FU6RD…pkYbeD
Sent
3 Nov 2026, 08:51 UTC
Source
e5885fc6c7d9

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 #88Done2 of 2 runs0.160×Guard failedGain lower bound 83.4%Guard failed
MLPvs #72Done2 of 2 runs0.269×Guard failedGain lower bound 69.8%Guard failed
ResNetvs #84Done2 of 2 runs0.154×QualifiesGain lower bound 84.0%Guard passed
MobileNetvs #85Done2 of 2 runs0.138×QualifiesGain lower bound 86.0%Guard passed
Normalized MLPvs #87Done2 of 2 runs0.391×Guard failedGain lower bound 58.2%Guard failed
Mixervs #86Done2 of 2 runs0.551×Guard failedGain lower bound 43.4%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 5611808390.154×Done1 attempt
  2. prod-arm64Seed 5611808390.166×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.088×
  • Batch 1, sparse0.086×
  • Batch 1, wide0.088×
  • Batch 8, normal0.260×
  • Batch 8, sparse0.255×
  • Batch 8, wide0.257×

Mixer

  • Batch 1, normal0.172×
  • Batch 1, sparse0.234×
  • Batch 1, wide1.200×
  • Batch 4, normal0.504×
  • Batch 4, sparse0.543×
  • Batch 4, wide0.587×

ResNet

  • Batch 1, normal0.084×
  • Batch 1, sparse0.103×
  • Batch 1, wide0.165×
  • Batch 4, normal0.100×
  • Batch 4, sparse0.111×
  • Batch 4, wide0.128×

MobileNet

  • Batch 1, normal0.085×
  • Batch 1, sparse0.099×
  • Batch 1, wide0.108×
  • Batch 4, normal0.092×
  • Batch 4, sparse0.101×
  • Batch 4, wide0.102×

Normalized MLP

  • Batch 1, normal0.105×
  • Batch 1, sparse0.128×
  • Batch 1, wide0.112×
  • Batch 8, normal0.217×
  • Batch 8, sparse0.228×
  • Batch 8, wide0.217×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 15846553780.269×Done1 attempt
  2. prod-arm64Seed 15846553780.269×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.067×
  • Batch 1, sparse0.069×
  • Batch 1, wide1.947×
  • Batch 8, normal0.210×
  • Batch 8, sparse0.214×
  • Batch 8, wide1.168×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 21963010820.168×Done1 attempt
  2. prod-arm64Seed 21963010820.141×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.110×
  • Batch 1, sparse0.131×
  • Batch 1, wide0.211×
  • Batch 4, normal0.129×
  • Batch 4, sparse0.142×
  • Batch 4, wide0.227×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 32485847130.149×Done1 attempt
  2. prod-arm64Seed 32485847130.127×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.119×
  • Batch 1, sparse0.141×
  • Batch 1, wide0.165×
  • Batch 4, normal0.126×
  • Batch 4, sparse0.135×
  • Batch 4, wide0.163×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 35791689150.392×Done1 attempt
  2. prod-arm64Seed 35791689150.389×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal0.112×
  • Batch 1, sparse0.141×
  • Batch 1, wide2.227×
  • Batch 8, normal0.298×
  • Batch 8, sparse0.309×
  • Batch 8, wide1.335×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 11561480430.544×Done1 attempt
  2. prod-arm64Seed 11561480430.559×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.162×
  • Batch 1, sparse0.228×
  • Batch 1, wide1.534×
  • Batch 4, normal0.474×
  • Batch 4, sparse0.521×
  • Batch 4, wide1.773×