Video-MME - Short Videos (w/o subs): leaderboard

Video-MME accuracy on short videos (under 2 minutes), without subtitles: four-option multiple-choice questions over videos from six visual domains, from the official leaderboard table. Accuracy in percent, higher is better.

Metric: Accuracy (%). Source: video-mme.github.io. Saturation forecast: Estimated already saturated. 51 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 1.5 Pro81.7#454
2GPT-4o80#357
3Gemini 1.5 Flash78.8#620
4GPT-4o Mini72.5#598
5MiniCPM-V-2.671.3#827
6Claude 3.5 Sonnet71#362
7Qwen-VL-Max55.8#475

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=video-mme-short-videos-w-o-subs · How It Works · Data refreshed daily, snapshot 2026-11-05.