Conjectures.io

Entry #112

Completed
Hotkey
5D1rds…AHt5bq
Sent
1 Nov 2026, 22:20 UTC
Source
a023e7a01b56

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 #54Done2 of 2 runs0.758×QualifiesGain lower bound 23.5%Guard passed
MLPvs #64Done2 of 2 runs1.209×Guard failedGain lower bound -22.8%Guard failed
ResNetvs #63Done2 of 2 runs1.033×Guard failedGain lower bound -5.5%Guard failed
MobileNetvs #62Done2 of 2 runs0.986×Guard failedGain lower bound -0.3%Guard failed
Normalized MLPvs #65Done2 of 2 runs1.206×Guard failedGain lower bound -21.7%Guard failed
Mixervs #61Done2 of 2 runs1.085×Guard failedGain lower bound -9.1%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 32897399520.745×Done1 attempt
  2. prod-arm64Seed 32897399520.771×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal0.885×
  • Batch 1, sparse0.862×
  • Batch 1, wide0.880×
  • Batch 8, normal0.397×
  • Batch 8, sparse0.666×
  • Batch 8, wide0.440×

Mixer

  • Batch 1, normal0.730×
  • Batch 1, sparse0.742×
  • Batch 1, wide0.746×
  • Batch 4, normal0.728×
  • Batch 4, sparse0.742×
  • Batch 4, wide0.749×

ResNet

  • Batch 1, normal0.857×
  • Batch 1, sparse0.895×
  • Batch 1, wide0.862×
  • Batch 4, normal0.847×
  • Batch 4, sparse0.886×
  • Batch 4, wide0.858×

MobileNet

  • Batch 1, normal0.874×
  • Batch 1, sparse0.899×
  • Batch 1, wide0.877×
  • Batch 4, normal0.875×
  • Batch 4, sparse0.896×
  • Batch 4, wide0.878×

Normalized MLP

  • Batch 1, normal0.875×
  • Batch 1, sparse0.859×
  • Batch 1, wide0.882×
  • Batch 8, normal0.462×
  • Batch 8, sparse0.704×
  • Batch 8, wide0.473×
MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 3988929251.323×Done1 attempt
  2. prod-arm64Seed 3988929251.105×Done1 attempt

By model and inputGuard at 1.100×

MLP

  • Batch 1, normal1.294×
  • Batch 1, sparse1.110×
  • Batch 1, wide1.333×
  • Batch 8, normal1.135×
  • Batch 8, sparse1.207×
  • Batch 8, wide1.239×
ResNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 23082115801.014×Done1 attempt
  2. prod-arm64Seed 23082115801.052×Done1 attempt

By model and inputGuard at 1.100×

ResNet

  • Batch 1, normal1.030×
  • Batch 1, sparse1.036×
  • Batch 1, wide1.033×
  • Batch 4, normal1.038×
  • Batch 4, sparse1.040×
  • Batch 4, wide1.038×
MobileNet2 runs

Runs

MachineTime
  1. prod-amd64Seed 12270115851.032×Done1 attempt
  2. prod-arm64Seed 12270115850.942×Done1 attempt

By model and inputGuard at 1.100×

MobileNet

  • Batch 1, normal0.968×
  • Batch 1, sparse0.991×
  • Batch 1, wide0.991×
  • Batch 4, normal0.978×
  • Batch 4, sparse0.992×
  • Batch 4, wide0.986×
Normalized MLP2 runs

Runs

MachineTime
  1. prod-amd64Seed 20196466621.295×Done1 attempt
  2. prod-arm64Seed 20196466621.123×Done1 attempt

By model and inputGuard at 1.100×

Normalized MLP

  • Batch 1, normal1.239×
  • Batch 1, sparse1.066×
  • Batch 1, wide1.257×
  • Batch 8, normal1.214×
  • Batch 8, sparse1.222×
  • Batch 8, wide1.240×
Mixer2 runs

Runs

MachineTime
  1. prod-amd64Seed 13889881831.118×Done1 attempt
  2. prod-arm64Seed 13889881831.053×Done1 attempt

By model and inputGuard at 1.100×

Mixer

  • Batch 1, normal1.121×
  • Batch 1, sparse1.069×
  • Batch 1, wide1.055×
  • Batch 4, normal1.096×
  • Batch 4, sparse1.067×
  • Batch 4, wide1.086×