AA MATH-500 — leaderboard
Artificial Analysis independent evaluation of MATH-500: 500 competition-level math problems across algebra, geometry, number theory, and more.
Metric: Accuracy (%). Source: artificialanalysis.ai. Status: saturated. 202 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 100 |
| 2 | GPT-5 (High) | 99.4 |
| 3 | O3 | 99.2 |
| 4 | GPT-5 (Medium) | 99.13 |
| 5 | Claude Sonnet 4 (Thinking) | 99.07 |
| 6 | Grok 4 | 99 |
| 7 | O4 Mini (High) | 98.87 |
| 8 | GPT-5 (Low) | 98.73 |
| 9 | Gemini 2.5 Pro (Preview 05-06) | 98.6 |
| 10 | O3 Mini (High) | 98.47 |
| 11 | Qwen 3 235B A22B 2507 (Thinking) | 98.4 |
| 12 | DeepSeek R1 0528 | 98.27 |
| 13 | Claude Opus 4 (Thinking) | 98.2 |
| 14 | Gemini 2.5 Flash (Thinking) | 98.13 |
| 15 | Qwen 3 235B A22B 2507 Instruct | 98 |
Interactive version: theaggregate.ai/benchmark?slug=aa-math-500 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.