A speaker plays music clearly next to a swimming pool. A person dips their head fully under the water, and the music drops to a faint distant rumble.
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 speaker
- a swimming pool
- a person
Event: a person submerges in a pool while a speaker plays music
Audio presence
- a speaker
Sound: music playback
Key standards
Visual physical commonsense (V-PC)
- A speaker is visibly present at the poolside with music playing from it.
- The person submerges into the pool with continuous head-going-fully-under motion.
Audio physical commonsense (A-PC)
- The music plays clearly above water from the visible speaker.
- The music drops to a faint distant rumble once the person is submerged.
Cross-modal physical commonsense (AV-PC)
- The shift to muffled distant music coincides temporally with the person's head submerging.