Conjectures.io

Entry #56

Completed
Hotkey
5Gq8EF…eUFyJj
Sent
30 Oct 2026, 10:10 UTC
Source
52dd96501c55

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 #32Done2 of 2 runs0.794×Guard failedGain lower bound 20.4%Guard failed
MLPvs #26Done2 of 2 runs1.300×Guard failedGain lower bound -31.7%Guard failed
ResNetvs #20Done2 of 2 runs0.968×QualifiesGain lower bound 2.9%Guard passed
MobileNetvs #31Done2 of 2 runs0.970×Guard failedGain lower bound 2.2%Guard failed
Normalized MLPvs #30Done2 of 2 runs1.195×Guard failedGain lower bound -20.9%Guard failed
Mixervs #27Done2 of 2 runs0.857×QualifiesGain lower bound 11.0%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 15237837480.767×Done1 attempt
  2. prod-arm64Seed 15237837480.822×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.900×
  • Batch 1, sparse1.025×
  • Batch 1, wide0.904×
  • Batch 8, normal0.872×
  • Batch 8, sparse0.791×
  • Batch 8, wide0.677×

Mixer

  • Batch 1, normal0.550×
  • Batch 1, sparse0.558×
  • Batch 1, wide0.549×
  • Batch 4, normal0.548×
  • Batch 4, sparse0.561×
  • Batch 4, wide0.550×

ResNet

  • Batch 1, normal0.913×
  • Batch 1, sparse0.917×
  • Batch 1, wide0.901×
  • Batch 4, normal0.924×
  • Batch 4, sparse0.929×
  • Batch 4, wide0.916×

MobileNet

  • Batch 1, normal0.897×
  • Batch 1, sparse0.908×
  • Batch 1, wide0.886×
  • Batch 4, normal0.912×
  • Batch 4, sparse0.920×
  • Batch 4, wide0.906×

Normalized MLP

  • Batch 1, normal0.856×
  • Batch 1, sparse0.932×
  • Batch 1, wide0.869×
  • Batch 8, normal0.812×
  • Batch 8, sparse0.736×
  • Batch 8, wide0.661×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 20877090071.293×Done1 attempt
  2. prod-arm64Seed 20877090071.307×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal1.081×
  • Batch 1, sparse1.284×
  • Batch 1, wide1.204×
  • Batch 8, normal1.551×
  • Batch 8, sparse1.306×
  • Batch 8, wide1.487×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 14593396770.969×Done1 attempt
  2. prod-arm64Seed 14593396770.967×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal0.963×
  • Batch 1, sparse0.973×
  • Batch 1, wide0.958×
  • Batch 4, normal0.971×
  • Batch 4, sparse0.984×
  • Batch 4, wide0.969×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 22260218080.956×Done1 attempt
  2. prod-arm64Seed 22260218080.985×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.896×
  • Batch 1, sparse1.076×
  • Batch 1, wide0.919×
  • Batch 4, normal0.916×
  • Batch 4, sparse1.099×
  • Batch 4, wide0.937×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 34810996491.213×Done1 attempt
  2. prod-arm64Seed 34810996491.177×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal1.044×
  • Batch 1, sparse1.212×
  • Batch 1, wide1.155×
  • Batch 8, normal1.384×
  • Batch 8, sparse1.076×
  • Batch 8, wide1.316×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 12410038990.868×Done1 attempt
  2. prod-arm64Seed 12410038990.847×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal0.951×
  • Batch 1, sparse0.937×
  • Batch 1, wide0.731×
  • Batch 4, normal0.919×
  • Batch 4, sparse0.913×
  • Batch 4, wide0.717×