EXAMPLE DATA — fictional examples, not real model performance / 当前为示例数据
Today’s monitoring alerts (2026-10-08): 1 models triggered M3 and other rules. Details →

Compare

Reference score 8.7

Duration: 18.6 s

Pelican likeness
Bicycle correctness
Riding pose
Composition
Obvious errors (higher is better)

(example) Clear pouch; chain missing.

Reference score 7.9

Duration: 9.1 s

Pelican likeness
Bicycle correctness
Riding pose
Composition
Obvious errors (higher is better)

(example) Composition OK; no grip.

Dimension comparison

Demo Pelican · 2026-09 · defaultExample Model · 2026-08 · default
Pelican likeness9.38.0
Bicycle correctness8.08.0
Riding pose9.08.0
Composition8.07.5
Obvious errors (higher is better)9.08.0