A ping-pong ball bounces on a wooden table with sharp ticks. Then drops onto carpet, and the bounces become soft and nearly silent.
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 ping-pong ball
- a wooden table
- carpet
Event: a ping-pong ball bounces on a surface
Audio presence
- a ping-pong ball
Sound: ball bounce impacts
Key standards
Visual physical commonsense (V-PC)
- A ping-pong-ball-sized ball bounces plausibly on the wooden table with higher rebounds and bounces on the carpet with lower rebounds.
Audio physical commonsense (A-PC)
- The table bounces have sharp bright ticks and the carpet bounces are distinctly softer and nearly silent.
Cross-modal physical commonsense (AV-PC)
- Each bounce sound coincides in time with a visible ball contact on its corresponding surface.
- The sharp bright ticks correspond to the ball bouncing on the wooden table.
- The soft, nearly silent bounces correspond to the ball landing on the carpet.