LLM Stats (VideoMME w sub.): leaderboard
Video-MME with subtitles: video understanding evaluation across short/medium/long videos with subtitle text provided as additional context.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Estimated already saturated. 12 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.8 Max | 90.4 | #35 |
| 2 | Seed 1.8 | 87.8 | #139 |
| 3 | Qwen 3.6 27B | 87.7 | #170 |
| 4 | Qwen 3.5 122B A10B | 87.3 | #173 |
| 5 | Qwen 3.5 27B | 87 | #187 |
| 6 | GPT-5 | 86.7 | #98 |
| 7 | Qwen 3.5 35B A3B | 86.6 | #256 |
| 8 | Qwen 3.6 35B A3B | 86.6 | #226 |
| 9 | Seed 2.0 Mini | 81.2 | #276 |
| 10 | Qwen 2.5 VL 32B Instruct | 77.9 | #453 |
| 11 | Qwen2.5-Omni-7B | 72.4 | #624 |
| 12 | Qwen 2.5 VL 7B Instruct | 71.6 | #654 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-videomme-w-sub · How It Works · Data refreshed daily, snapshot 2026-10-20.