Molmo2 Human Preference - Captioning: leaderboard
Metric: Elo rating (bootstrap median, Bradley-Terry; pairwise human preferences on dense captions of 51 videos). Source: arxiv.org. Saturation forecast: Around February 2027. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 1148 |
| 2 | GPT-5 | 1136 |
| 3 | Gemini 3 Pro | 1126 |
| 4 | Gemini 2.5 Flash | 1109 |
| 5 | Qwen 3 VL 8B | 1105 |
| 6 | GPT-5 Mini | 1086 |
| 7 | Qwen 3 VL 4B | 1052 |
| 8 | Molmo2-8B | 1049 |
| 9 | GLM-4.1V-9B (Thinking) | 1013 |
| 10 | Claude Sonnet 4.5 | 1009 |
Interactive version: theaggregate.ai/benchmark?slug=molmo2-human-preference-captioning · How It Works · Data refreshed daily, snapshot 2026-09-25.