OTIS Mock AIME 2024-2025: leaderboard
Competition-level math problems from OTIS Mock AIME evaluating olympiad-level math.
Metric: Score (%). Source: neosignal.io. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 96.11 | #110 |
| 2 | Gemini 3 Pro | 92.78 | #80 |
| 3 | GPT-OSS-120B | 88.89 | #341 |
| 4 | DeepSeek V3 | 87.82 | #321 |
| 5 | Qwen 3 235B A22B | 86.67 | #313 |
| 6 | Claude Opus 4.5 | 86.11 | #83 |
| 7 | Gemini 2.5 Pro (Preview 06-05) | 84.72 | #106 |
| 8 | Grok 4 | 84 | #181 |
| 9 | O3 | 83.89 | #124 |
| 10 | Kimi K2 (Thinking) | 83.06 | #243 (Kimi K2) |
| 11 | O4 Mini (High) | 81.67 | #180 (O4 Mini) |
| 12 | Claude Sonnet 4.5 | 77.78 | #142 |
| 13 | Grok 3 Mini | 77.78 | #281 |
| 14 | O1 | 73.33 | #183 |
| 15 | Qwen 3 Max Instruct | 73.33 | #259 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=otis-mock-aime-2024-2025 · How It Works · Data refreshed daily, snapshot 2026-10-20.