Conjectures.io

Entry #152

Completed
Hotkey
5Dz387…gKuj9r
Sent
3 Nov 2026, 16:56 UTC
Source
fa61b3b92360

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 #99Done2 of 2 runs0.829×QualifiesGain lower bound 15.6%Guard passed
MLPvs #104Done2 of 2 runs0.975×Above the barGain lower bound 2.3%Guard passed
ResNetvs #101Done2 of 2 runs0.836×QualifiesGain lower bound 14.5%Guard passed
MobileNetvs #105Done2 of 2 runs0.903×QualifiesGain lower bound 9.1%Guard passed
Normalized MLPvs #106Done2 of 2 runs0.945×QualifiesGain lower bound 5.1%Guard passed
Mixervs #107Done2 of 2 runs0.880×QualifiesGain lower bound 11.6%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 24073688550.829×Done1 attempt
  2. prod-arm64Seed 24073688550.828×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.880×
  • Batch 1, sparse0.882×
  • Batch 1, wide0.882×
  • Batch 8, normal0.978×
  • Batch 8, sparse0.978×
  • Batch 8, wide0.978×

Mixer

  • Batch 1, normal0.783×
  • Batch 1, sparse0.781×
  • Batch 1, wide0.779×
  • Batch 4, normal0.675×
  • Batch 4, sparse0.660×
  • Batch 4, wide0.654×

ResNet

  • Batch 1, normal0.843×
  • Batch 1, sparse0.843×
  • Batch 1, wide0.832×
  • Batch 4, normal0.863×
  • Batch 4, sparse0.843×
  • Batch 4, wide0.845×

MobileNet

  • Batch 1, normal0.781×
  • Batch 1, sparse0.757×
  • Batch 1, wide0.769×
  • Batch 4, normal0.812×
  • Batch 4, sparse0.811×
  • Batch 4, wide0.827×

Normalized MLP

  • Batch 1, normal0.814×
  • Batch 1, sparse0.813×
  • Batch 1, wide0.817×
  • Batch 8, normal0.943×
  • Batch 8, sparse0.956×
  • Batch 8, wide0.947×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 32921894870.978×Done1 attempt
  2. prod-arm64Seed 32921894870.972×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.956×
  • Batch 1, sparse0.956×
  • Batch 1, wide0.958×
  • Batch 8, normal0.993×
  • Batch 8, sparse0.993×
  • Batch 8, wide0.992×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 9139559450.822×Done1 attempt
  2. prod-arm64Seed 9139559450.850×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.851×
  • Batch 1, sparse0.851×
  • Batch 1, wide0.858×
  • Batch 4, normal0.800×
  • Batch 4, sparse0.846×
  • Batch 4, wide0.847×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 25455644820.881×Done1 attempt
  2. prod-arm64Seed 25455644820.925×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.905×
  • Batch 1, sparse0.902×
  • Batch 1, wide0.896×
  • Batch 4, normal0.909×
  • Batch 4, sparse0.907×
  • Batch 4, wide0.906×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 12074297120.942×Done1 attempt
  2. prod-arm64Seed 12074297120.948×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal0.922×
  • Batch 1, sparse0.921×
  • Batch 1, wide0.919×
  • Batch 8, normal0.964×
  • Batch 8, sparse0.978×
  • Batch 8, wide0.968×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 40632287670.874×Done1 attempt
  2. prod-arm64Seed 40632287670.885×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.890×
  • Batch 1, sparse0.893×
  • Batch 1, wide0.891×
  • Batch 4, normal0.880×
  • Batch 4, sparse0.877×
  • Batch 4, wide0.879×