Math-PT (pt-BR MC) - Level 2: leaderboard

Math-PT (2026) tests mathematical reasoning on problems written natively in Brazilian Portuguese; this board covers the level-2 multiple-choice questions from OBMEP. Accuracy in percent on the boxed answer letter, higher is better.

Metric: Accuracy (%) on the 118 level-2 multiple-choice questions in Brazilian Portuguese (OBMEP); 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: Not forecast. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1GPT-590.68#185
2Qwen 3 14B90.68#1284
3Qwen 3 32B88.98#983
4Qwen 3 235B A22B88.98#633
5Qwen 3 8B87.29#1890
6Gemini 2.5 Flash83.9#466
7Claude Haiku 4.583.9#587
8DeepSeek V3.183.05#511
9Qwen 3 30B A3B82.2#1165
10Gemma 3 27B (IT)66.95#1341
11Llama 3.3 70B Instruct35.59#1394

No result here: #2 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-2 · How It Works · Data refreshed daily, snapshot 2026-11-22.