VURB - General Video Understanding: leaderboard

Metric: Majority-vote pairwise accuracy (%; eight judgments with the response order shuffled, a pair counts when the chosen response wins more than four, over pairs with a valid verdict; 797 pairs from VideoMME short and medium, MotionBench and MVBench). Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.

Top models

#ModelScore
1GPT-5.262.6
2Qwen 3 VL 32B (Thinking)54.5
3Qwen 3 VL 8B Instruct51.3
4Qwen 3 VL 8B (Thinking)50.4

Interactive version: theaggregate.ai/benchmark?slug=vurb-general-video-understanding · How It Works · Data refreshed daily, snapshot 2026-09-25.