A street musician plays an accordion on a sidewalk. A listener walks slowly away and the music grows steadily quieter with each step.
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 street musician
- an accordion
- a listener
- a sidewalk
Event: a musician plays an accordion while a listener walks on a sidewalk
Audio presence
- an accordion
Sound: accordion music
Key standards
Visual physical commonsense (V-PC)
- The accordion player continues playing the accordion while a listener visibly walks farther away along the sidewalk during the clip.
Audio physical commonsense (A-PC)
- Accordion music is heard throughout the segment in which the player is visibly performing, and that music grows steadily quieter as the clip progresses.
Cross-modal physical commonsense (AV-PC)
- The decrease in accordion loudness corresponds to the visibly increasing distance between the listener and the player.