Conjectures.io

Entry #159

Completed
Hotkey
5FEnJz…2aPoH6
Sent
4 Nov 2026, 02:26 UTC
Source
640530867bcb

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 runs0.841×Guard failedGain lower bound 15.7%Guard failed
MLPvs #113Done2 of 2 runs0.982×Above the barGain lower bound 1.4%Guard passed
ResNetvs #115Done2 of 2 runs0.962×Guard failedGain lower bound 3.3%Guard failed
MobileNetvs #117Done2 of 2 runs0.957×QualifiesGain lower bound 3.6%Guard passed
Normalized MLPvs #114Done2 of 2 runs0.976×Above the barGain lower bound 2.2%Guard passed
Mixervs #116Done2 of 2 runs0.914×QualifiesGain lower bound 7.8%Guard passed

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 21357888020.853×Done1 attempt
  2. prod-arm64Seed 21357888020.830×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.840×
  • Batch 1, sparse0.853×
  • Batch 1, wide0.854×
  • Batch 8, normal0.834×
  • Batch 8, sparse0.843×
  • Batch 8, wide0.843×

Mixer

  • Batch 1, normal0.888×
  • Batch 1, sparse0.871×
  • Batch 1, wide0.886×
  • Batch 4, normal0.808×
  • Batch 4, sparse0.786×
  • Batch 4, wide0.809×

ResNet

  • Batch 1, normal0.903×
  • Batch 1, sparse0.912×
  • Batch 1, wide0.890×
  • Batch 4, normal0.791×
  • Batch 4, sparse0.761×
  • Batch 4, wide0.759×

MobileNet

  • Batch 1, normal0.898×
  • Batch 1, sparse0.889×
  • Batch 1, wide0.923×
  • Batch 4, normal0.755×
  • Batch 4, sparse0.797×
  • Batch 4, wide0.771×

Normalized MLP

  • Batch 1, normal0.862×
  • Batch 1, sparse0.863×
  • Batch 1, wide0.893×
  • Batch 8, normal0.863×
  • Batch 8, sparse0.846×
  • Batch 8, wide0.855×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 40706440650.990×Done1 attempt
  2. prod-arm64Seed 40706440650.975×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.970×
  • Batch 1, sparse0.970×
  • Batch 1, wide0.970×
  • Batch 8, normal0.996×
  • Batch 8, sparse0.995×
  • Batch 8, wide0.996×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 14652053270.966×Done1 attempt
  2. prod-arm64Seed 14652053270.958×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.984×
  • Batch 1, sparse1.005×
  • Batch 1, wide0.952×
  • Batch 4, normal0.966×
  • Batch 4, sparse0.931×
  • Batch 4, wide0.948×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 11168594150.925×Done1 attempt
  2. prod-arm64Seed 11168594150.991×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.966×
  • Batch 1, sparse0.964×
  • Batch 1, wide0.962×
  • Batch 4, normal0.947×
  • Batch 4, sparse0.939×
  • Batch 4, wide0.936×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 40946192610.982×Done1 attempt
  2. prod-arm64Seed 40946192610.970×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal0.958×
  • Batch 1, sparse0.961×
  • Batch 1, wide0.962×
  • Batch 8, normal0.991×
  • Batch 8, sparse0.991×
  • Batch 8, wide0.991×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 31596450710.919×Done1 attempt
  2. prod-arm64Seed 31596450710.909×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.933×
  • Batch 1, sparse0.934×
  • Batch 1, wide0.934×
  • Batch 4, normal0.882×
  • Batch 4, sparse0.899×
  • Batch 4, wide0.897×