A snake opens its mouth slightly in a close-up shot, yet the sound is natural human laughter emerging directly from the snake, despite the animal lacking the anatomy to produce such a voice.
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 snake
Event: a snake opens its mouth slightly in close-up
Audio presence
- a snake
Sound: human laughter
Key standards
Visual physical commonsense (V-PC)
- A clearly identifiable snake is visible in close-up and visibly opens and partially closes its mouth in continuous physically plausible motion.
- The animal in the frame retains its snake anatomy, scales, and elongated body throughout the clip without morphing into a human or other species.
Audio physical commonsense (A-PC)
- The audio is natural human laughter with the pitch, rhythm, and breath pattern of a real laughing person.
- The audio contains no snake hiss or any reptilian sound at any point.
Cross-modal physical commonsense (AV-PC)
- The laughter sound emerges in time with the snake's visible mouth motion, with the sound originating from the visible snake as the only on-screen sound source.