A small firecracker pops with a sharp high crack. A large firework shell explodes with a deep thunderous boom.
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 firecracker
- a firework shell
Event: a firecracker and a firework shell explode
Audio presence
- a firecracker
- a firework shell
Sound: ['a firecracker crack', 'a firework shell boom']
Key standards
Visual physical commonsense (V-PC)
- Both the small firecracker and the large firework shell visibly detonate in the clip, each producing its own distinct explosion.
Audio physical commonsense (A-PC)
- The large firework shell's boom is noticeably lower in pitch than the small firecracker's sharp crack.
- The sounds are discrete real explosion impulses.
Cross-modal physical commonsense (AV-PC)
- Each explosion sound aligns in time with the corresponding visible detonation.
- The sharp high crack corresponds to the small firecracker.
- The deep thunderous boom corresponds to the large firework shell.