A child pops a balloon at a birthday party. Nearby guests jump at the sharp pop. The scene cuts to the other end of the large hall, where the same pop sounds like a faint distant snap.
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 child
- a balloon
- guests
- a hall
Event: a child pops a balloon in a hall
Audio presence
- a balloon
Sound: a balloon pop
Key standards
Visual physical commonsense (V-PC)
- A near-scene shows a child popping a balloon at a birthday party, and a far-scene from the opposite end of the same hall is also shown.
Audio physical commonsense (A-PC)
- A balloon-pop sound is heard in both scenes, and the far-scene pop is noticeably quieter and thinner than the near-scene pop.
Cross-modal physical commonsense (AV-PC)
- Each pop sound aligns in time with the visible balloon-popping moment in its scene.