MathNet - Text-Only Problems: leaderboard

MathNet (2026) is a multilingual set of Olympiad problems with official solutions from national competition booklets of 47 countries; on its problem-solving test set a GPT-5 grader scores each solution from 0 to 7 against the official one, and 6 or more counts as correct. Accuracy on the problems that have no figures, so every model sees the same input, in percent; higher is better.

Metric: Problem-solving accuracy (%) on the problems without images on MathNet-Solve-Test, Olympiad problems from official national competition booklets of 47 countries; a GPT-5 grader scores each solution 0-7 against the official solution and 6 or more counts as correct; image-capable models get the figures, text-only models a text description of them; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 14 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3.1 Pro (Preview)76.7#102
2GPT-568.6#185
3Gemini 3 Flash (Preview)67.8#156
4GPT-5 Mini55.6#373
5GPT-5 Nano45.1#912
6Claude Opus 4.6Thinking43.8#114
7DeepSeek V3.2Non-reasoning40.4#425
8Gemini 2.5 Flash38#466
9MathNet DeepSeek-R1 (checkpoint unspecified)35.4
10Grok 327#614
11GPT-4.120.8#506
12Llama 4 Maverick Instruct FP814.1#1034
13GPT-4o6#754
14Ministral 3B3#6849

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mathnet-text-only-problems · How It Works · Data refreshed daily, snapshot 2026-11-22.