A man wearing boots walks across fresh snow in daylight; the sound is wet mud squelching, not crisp snow compression.
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
- boots
- fresh snow
Event: a man wearing boots walks across fresh snow in daylight
Audio presence
- boots
- fresh snow
Sound: wet mud squelching
Key standards
Visual physical commonsense (V-PC)
- A man wearing boots walks across visibly fresh snow in daylight, with each footstep visibly compressing the snow surface.
- The ground surface remains visibly snow-covered throughout the clip, with no visible mud, water, or wet ground at any footstep point.
Audio physical commonsense (A-PC)
- Each footstep sound is a wet, sloppy mud squelch with the moist viscous character of stepping into wet mud.
Cross-modal physical commonsense (AV-PC)
- Each squelch sound aligns in time with a visible foot strike on the snow surface.