MathArena - USAMO 2025 — leaderboard
Metric: Accuracy (%). Source: matharena.ai. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 100 |
| 2 | DeepSeek R1 0528 | 30.06 |
| 3 | Gemini 2.5 Pro | 24.4 |
| 4 | O4 Mini (High) | 19.05 |
| 5 | DeepSeek R1 | 4.76 |
| 6 | Gemini 2.0 Flash (Thinking) | 4.17 |
| 7 | Claude 3.7 Sonnet (Thinking) | 3.65 |
| 8 | QwQ-32B | 2.98 |
| 9 | O3 Mini (High) | 2.08 |
| 10 | Median Human | 0 |
Interactive version: theaggregate.ai/benchmark?slug=matharena-usamo-2025 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.