Math-PT (pt-PT OE) - Level 2: leaderboard

Math-PT (2026) tests mathematical reasoning on problems written natively in European Portuguese; this board covers the level-2 open-ended questions from OPM. 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 96 level-2 open-ended questions in European Portuguese (OPM); 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

#ModelScoreOverall rank
1GPT-589.58#178
2Qwen 3 235B A22B79.17#628
3Qwen 3 14B73.96#1269
4Gemini 2.5 Flash72.92#463
5Qwen 3 32B70.83#972
6DeepSeek V3.167.71#511
7Qwen 3 30B A3B64.58#1156
8Qwen 3 8B58.33#1873
9Claude Haiku 4.547.92#583
10Gemma 3 27B (IT)30.21#1324
11Llama 3.3 70B Instruct17.71#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-pt-oe-level-2 · How It Works · Data refreshed daily, snapshot 2026-11-21.