AIME 2025: leaderboard
All 30 problems from the 2025 American Invitational Mathematics Examination, testing olympiad-level mathematical reasoning.
Metric: Accuracy (%). Source: artificialanalysis.ai. Status: saturated. 178 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Best Expert | 100 |
| 2 | GPT-5 Codex (High) | 98.67 |
| 3 | MiMo-V2-Flash (Reasoning) | 96.33 |
| 4 | Gemini 3 Pro (Preview) (High) | 95.67 |
| 5 | GLM-4.7 (Reasoning) | 95 |
| 6 | Kimi K2 (Thinking) | 94.67 |
| 7 | GPT-5 (High) | 94.33 |
| 8 | GPT-5.1 (High) | 94 |
| 9 | GPT-OSS-120B (High) | 93.44 |
| 10 | Grok 4 | 92.67 |
| 11 | GPT-5 (Medium) | 91.67 |
| 12 | Claude Opus 4.5 (Thinking) | 91.33 |
| 13 | Qwen 3 235B A22B 2507 (Thinking) | 91 |
| 14 | GPT-5 Mini (High) | 90.67 |
| 15 | Grok 4 Fast (Reasoning) | 89.67 |
Interactive version: theaggregate.ai/benchmark?slug=aime-2025 · How It Works · Data refreshed daily, snapshot 2026-09-05.