A man shouts loudly in a bedroom. He buries his face into a pillow and continues shouting, his voice now a soft muffled cry.
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 man
- a pillow
- a bedroom
Event: a man shouts and then buries his face into a pillow
Audio presence
- a man
Sound: a man shouting
Key standards
Visual physical commonsense (V-PC)
- A man shouts in a bedroom with his face raised away from the pillow.
- The same man then shouts again with his face buried into the pillow.
Audio physical commonsense (A-PC)
- The first shout is loud and bright in timbre.
- The second shout is the same man's voice still performing a shout, softened by the pillow.
- The second shout is noticeably quieter and duller in timbre than the first shout.
Cross-modal physical commonsense (AV-PC)
- The transition from a clear bright shout to a softer muffled shout aligns in time with the moment the man's face contacts the pillow.