A person drops a marble onto a hardwood floor. The impact is heard clearly by a person in the room below through the solid floor.
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 marble
- a hardwood floor
- two people
Event: a marble is dropped onto a hardwood floor with a listener in the room below
Audio presence
- a marble
- a hardwood floor
Sound: marble impact transmitted through floor
Key standards
Visual physical commonsense (V-PC)
- A marble is dropped onto an upstairs hardwood floor, and a separate downstairs room with a listener is also visible.
- The upstairs and downstairs rooms have geometrically plausible floor heights, and the marble stays within the upstairs room.
Audio physical commonsense (A-PC)
- The sound heard downstairs is a conducted marble-on-hardwood thud, plausibly attenuated through the floor.
Cross-modal physical commonsense (AV-PC)
- The downstairs impact sound onsets at the moment the marble visibly strikes the upstairs floor.