Math-PT (pt-BR MC) - Level 4: leaderboard
Math-PT (2026) tests mathematical reasoning on problems written natively in Brazilian Portuguese; this board covers the level-4 multiple-choice questions from OBMEP and OMIF. Accuracy in percent on the boxed answer letter, higher is better.
Metric: Accuracy (%) on the 270 level-4 multiple-choice questions in Brazilian Portuguese (OBMEP and OMIF); native-language zero-shot prompt asking for a boxed final answer, figures given as LaTeX source; the letter in the final boxed answer must match the key; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 93.33 | #178 |
| 2 | Qwen 3 235B A22B | 91.48 | #628 |
| 3 | DeepSeek V3.1 | 89.26 | #511 |
| 4 | Qwen 3 14B | 88.15 | #1269 |
| 5 | Gemini 2.5 Flash | 85.93 | #463 |
| 6 | Qwen 3 32B | 85.19 | #972 |
| 7 | Qwen 3 30B A3B | 84.44 | #1156 |
| 8 | Claude Haiku 4.5 | 81.48 | #583 |
| 9 | Qwen 3 8B | 81.48 | #1873 |
| 10 | Gemma 3 27B (IT) | 64.81 | #1324 |
| 11 | Llama 3.3 70B Instruct | 32.96 | #1385 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=math-pt-pt-br-mc-level-4 · How It Works · Data refreshed daily, snapshot 2026-11-21.