A person cracks an egg on the edge of a bowl. The shell snaps crisply, and the egg slides out with a soft wet slosh.
Seedance 2.0ByteDance · proprietary
Kling 3.0 OmniKuaishou · proprietary
Veo 3.1Google · proprietary
LTX-2.3Lightricks · open
Ovi 1.1character.ai · open
JavisDiT++JavisDiT · open
MagiHumanGAIR-NLP · open
Per-prompt rubric
Basic standards
Visual presence
- a person
- an egg
- a bowl
Event: an egg is cracked on the edge of a bowl
Audio presence
- an eggshell
- raw egg
Sound: ['eggshell crack', 'wet egg slosh']
Key standards
Visual physical commonsense (V-PC)
- The egg shell visibly breaks open at the moment of contact with the bowl edge, and the egg contents pour out with plausible viscous fluid behavior.
- Only a single egg's contents come out of the shell.
Audio physical commonsense (A-PC)
- The shell crack has a sharp brittle timbre.
- The egg pour has a soft wet slosh timbre.
Cross-modal physical commonsense (AV-PC)
- The crack sound onset aligns in time with the visible contact between the egg and the bowl edge.
- The slosh sound aligns in time with the visible pour of egg contents into the bowl.