A person sits inside a quiet café. The front door swings open and loud street noise floods in for a moment. Then the door closes and the café returns to quiet.
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
- a cafe front door
- a cafe interior
Event: a cafe front door swings open and then closes
Audio presence
- street noise
Sound: street noise
Key standards
Visual physical commonsense (V-PC)
- A person sits inside a cafe interior while the cafe front door visibly opens and then closes.
- The door motion is continuous and the cafe geometry remains stable throughout the clip.
Audio physical commonsense (A-PC)
- Street noise is quiet inside the cafe while the door is closed.
- Street noise becomes loud and broadband while the door is open.
- Street noise drops back to a quiet residual after the door closes again.
Cross-modal physical commonsense (AV-PC)
- The surges and drops in street noise align in time with the visible opening and closing of the cafe door.