AA AIME 2025 — leaderboard
Artificial Analysis independent evaluation of all 30 problems from the 2025 American Invitational Mathematics Examination (olympiad-level).
Metric: Accuracy (%). Source: artificialanalysis.ai. Status: saturated. 269 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Human Expert | 100 |
| 2 | GPT-5.2 (xHigh) | 99 |
| 3 | GPT-5 Codex (High) | 98.67 |
| 4 | GPT-5.2 (Medium) | 96.67 |
| 5 | DeepSeek V3.2 Speciale | 96.67 |
| 6 | MiMo-V2-Flash (Reasoning) | 96.33 |
| 7 | Gemini 3 Pro (Preview) (High) | 95.67 |
| 8 | GPT-5.1 Codex (High) | 95.67 |
| 9 | GLM-4.7 (Reasoning) | 95 |
| 10 | Kimi K2 (Thinking) | 94.67 |
| 11 | KAT-Coder-Pro V1 | 94.67 |
| 12 | GPT-5 (High) | 94.33 |
| 13 | GPT-5.1 (High) | 94 |
| 14 | GPT-OSS-120B (High) | 93.44 |
| 15 | Grok 4 | 92.67 |
Interactive version: theaggregate.ai/benchmark?slug=aa-aime-2025 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.