VMMU - Geography: leaderboard
Metric: Accuracy (%) on the VMMU Geography 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 | 94.59 |
| 2 | Claude Opus 4.6 (Thinking) | 90.85 |
| 3 | GPT-5 (Medium) | 89.4 |
| 4 | O3 (Medium) | 88.98 |
| 5 | Gemini 2.5 Flash | 85.24 |
| 6 | Qwen 2.5 VL 32B Instruct | 71.31 |
| 7 | GPT-4.1 | 69.85 |
| 8 | Llama 4 Scout | 69.02 |
| 9 | Qwen 2.5 VL 72B Instruct | 64.66 |
| 10 | Llama 4 Maverick | 53.22 |
| 11 | Claude Sonnet 4 | 48.44 |
| 12 | Gemma 3 27B | 47.4 |
| 13 | Mistral Medium 3 | 36.59 |
| 14 | Mistral Small 3.2 | 30.98 |
| 15 | Gemma 3 4B | 27.65 |
Interactive version: theaggregate.ai/benchmark?slug=vmmu-geography · How It Works · Data refreshed daily, snapshot 2026-09-29.