A person claps gently, producing a soft pat. Then claps with full force, producing a loud sharp crack.
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
- hands
Event: a person claps hands
Audio presence
- hands
Sound: a hand clap
Key standards
Visual physical commonsense (V-PC)
- The person claps their hands at least twice, first gently and then with clearly more force, with the two hands visibly meeting on each clap.
Audio physical commonsense (A-PC)
- The forceful clap is noticeably louder than the gentle clap.
- Each clap sound is a real hand-on-hand clap.
Cross-modal physical commonsense (AV-PC)
- Each clap sound aligns in time with the visible contact of the two hands.
- The louder clap corresponds to the forceful hand contact.
- The quieter clap corresponds to the gentle hand contact.