DynaMath: leaderboard
A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.
Metric: Score (%). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2 VL 72B | 55.1 | #460 |
| 2 | Llama 3.2 90B | 44 | #596 |
| 3 | Qwen 2 VL 7B | 42.1 | #750 |
| 4 | InternVL2-8B | 39.7 | #817 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=dynamath · How It Works · Data refreshed daily, snapshot 2026-10-20.