VMMU - Mathematics: leaderboard
Metric: Accuracy (%) on the VMMU Mathematics questions (Vietnamese page-image multiple choice); higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 92.54 |
| 2 | GPT-5 (Medium) | 91.23 |
| 3 | O3 (Medium) | 84.87 |
| 4 | Claude Opus 4.6 (Thinking) | 83.33 |
| 5 | Gemini 2.5 Flash | 82.46 |
| 6 | Claude Sonnet 4 | 64.25 |
| 7 | Qwen 2.5 VL 32B Instruct | 61.18 |
| 8 | Llama 4 Scout | 59.65 |
| 9 | Qwen 2.5 VL 72B Instruct | 58.33 |
| 10 | Gemma 3 27B | 47.37 |
| 11 | GPT-4.1 | 46.27 |
| 12 | Mistral Medium 3 | 44.74 |
| 13 | Llama 4 Maverick | 42.54 |
| 14 | Mistral Small 3.2 | 34.65 |
| 15 | Gemma 3 4B | 26.32 |
Interactive version: theaggregate.ai/benchmark?slug=vmmu-mathematics · How It Works · Data refreshed daily, snapshot 2026-09-29.