A smartphone lies on a bed playing music at a steady volume in a quiet bedroom; initially, the phone is fully uncovered and the music sounds normal; a thick pillow is then placed directly on top of the phone, completely covering the speaker; the sound becomes louder, clearer, and brighter after being covered, with increased high-frequency detail.
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 smartphone
- a pillow
- a bed
Event: a thick pillow is placed on top of a smartphone playing music on a bed
Audio presence
- a smartphone
Sound: music from a smartphone
Key standards
Visual physical commonsense (V-PC)
- A smartphone lies fully uncovered on a bed at the start of the clip, visibly playing music with no pillow on top.
- A thick pillow is then placed directly on top of the smartphone, completely covering the phone speaker for the rest of the clip.
Audio physical commonsense (A-PC)
- The music in the uncovered phase plays at a steady volume with the typical small-speaker character of phone audio.
- The music in the covered phase becomes noticeably louder, clearer, and brighter than in the uncovered phase, with increased high-frequency detail.
Cross-modal physical commonsense (AV-PC)
- The shift to louder clearer brighter music aligns in time with the moment the pillow visibly covers the smartphone speaker.