AoPS Strategy Coverage - Number Theory: leaderboard
The number theory problems of the AoPS strategy-coverage evaluation (Yang et al., 2026): the share of the 45 human solution-strategy families from Art of Problem Solving that a model recovers when asked for multiple distinct solutions per problem. Percent; higher is better.
Metric: AoPS-reference strategies recovered (% of the 45 reference strategy families for the 15 number theory problems; multiple-strategy prompt). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 68.89 | #91 |
| 2 | DeepSeek V3.2 (Thinking) | 64.44 | #403 (DeepSeek V3.2) |
| 3 | GPT-5.4 | 62.22 | #121 |
| 4 | Claude Opus 4.6 | 42.22 | #100 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=aops-strategy-coverage-number-theory · How It Works · Data refreshed daily, snapshot 2026-11-12.