A single person standing in frame raises their hands to clap; the full clap sound and its echo are heard clearly before the hands visually meet, and when the hands finally touch there is no sound. This intentionally violates audio–visual causality, with sound arriving before motion.
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
- two hands
Event: a person raises both hands and brings them together in a clap
Audio presence
- a person
Sound: a hand clap and its echo
Key standards
Visual physical commonsense (V-PC)
- A single person stands in frame and visibly raises both hands toward each other in a continuous clap motion that culminates in palm contact.
Audio physical commonsense (A-PC)
- The audio contains a clear hand clap with the sharp percussive slap of two human palms, followed by a discrete echo with reduced amplitude.
Cross-modal physical commonsense (AV-PC)
- The clap sound and its echo play in the audio before the hands visibly meet in the frame.
- The audio is silent at the moment the hands visibly make palm contact, with no clap onset coinciding with the contact.