AMVICC - MMVP Pair Accuracy - Structural and Physical Characteristics: leaderboard
Metric: Pair accuracy (%; the 15 MMVP image pairs in the Structural and Physical Characteristics category, a pair credited only when both of its questions are answered correctly; free-form answers graded by GPT-4 against the authors' rubric). Source: arxiv.org. Saturation forecast: Around March 2027. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.2 90B Vision Instruct | 93.33 |
| 2 | Llama 4 Maverick | 86.67 |
| 3 | Llama 4 Scout | 86.67 |
| 4 | Qwen 2.5 VL 72B Instruct | 73.33 |
| 5 | pixtral-large-2411 | 73.33 |
| 6 | GPT-4o | 66.67 |
| 7 | Gemini 2.5 Pro | 60 |
| 8 | Claude Sonnet 4 | 60 |
| 9 | Claude Opus 4.1 | 60 |
| 10 | Grok 4 | 53.33 |
| 11 | Gemma 3 27B | 46.67 |
Interactive version: theaggregate.ai/benchmark?slug=amvicc-mmvp-pair-accuracy-structural-and-physical-characteristics · How It Works · Data refreshed daily, snapshot 2026-09-29.