A pianist strikes a key without the sustain pedal and the note fades quickly. They strike the same key with the pedal pressed and the note rings on much longer.
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 pianist
- a piano
- a sustain pedal
Event: a pianist strikes a piano key
Audio presence
- a piano
Sound: a piano note
Key standards
Visual physical commonsense (V-PC)
- A pianist visibly strikes a piano key without pressing the sustain pedal during the first part of the clip.
- The pianist visibly strikes a piano key with the sustain pedal pressed down by foot during the second part of the clip.
Audio physical commonsense (A-PC)
- The first piano note fades quickly with rapid decay.
- The second piano note rings on much longer with sympathetic resonance from other strings.
Cross-modal physical commonsense (AV-PC)
- Each piano note onset aligns in time with the visible key strike.
- The longer-decaying note in the audio is the one whose strike aligns with the pedal-pressed state.