Math-PT (pt-BR OE) - Level 2: leaderboard
Math-PT (2026) tests mathematical reasoning on problems written natively in Brazilian Portuguese; this board covers the level-2 open-ended questions from OBM. Accuracy in percent, with Kimi K2 Thinking checking that the final answer is mathematically equivalent to the gold answer, higher is better.
Metric: Accuracy (%) on the 32 level-2 open-ended questions in Brazilian Portuguese (OBM); native-language zero-shot prompt asking for a boxed final answer, figures given as LaTeX source; Kimi K2 Thinking judges whether the final answer is mathematically equivalent to the gold answer; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 90.62 | #178 |
| 2 | Gemini 2.5 Flash | 84.38 | #463 |
| 3 | Qwen 3 32B | 84.38 | #972 |
| 4 | Qwen 3 235B A22B | 84.38 | #628 |
| 5 | Qwen 3 14B | 81.25 | #1269 |
| 6 | DeepSeek V3.1 | 78.12 | #511 |
| 7 | Qwen 3 8B | 75 | #1873 |
| 8 | Qwen 3 30B A3B | 75 | #1156 |
| 9 | Claude Haiku 4.5 | 68.75 | #583 |
| 10 | Gemma 3 27B (IT) | 65.62 | #1324 |
| 11 | Llama 3.3 70B Instruct | 34.38 | #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-oe-level-2 · How It Works · Data refreshed daily, snapshot 2026-11-21.