A person counts from one to ten facing the camera. Each spoken number matches the mouth shape exactly.
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 person
Event: a person faces the camera and moves their mouth to speak
Audio presence
- a person
Sound: a counting voice
Key standards
Visual physical commonsense (V-PC)
- A visible person faces the camera and produces physically plausible mouth and lip movements consistent with speech.
Audio physical commonsense (A-PC)
- The voice is the natural timbre of a human speaking voice counting the distinct numbers from one through ten in order.
Cross-modal physical commonsense (AV-PC)
- Each spoken number aligns in time with the corresponding visible mouth shape.
- The lip-sync remains accurate across the full count from one through ten without drift.