Benchmark
AI video model leaderboard
Every model runs the same 40 prompts. A vision model scores each clip
blind against the original prompt, three independent passes, across visual fidelity, physics and
motion, subject consistency, prompt adherence and audio sync. Scores are out of 5.
Full methodology.
No scored runs yet.
The prompt battery and the model list are loaded. Scores land here once the first round has been generated and judged.