LLM Stats (AIME 2024): leaderboard
Problems from the 2024 American Invitational Mathematics Examination testing advanced math reasoning across algebra, combinatorics, number theory, and geometry.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Not forecast. 54 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Best Expert | 100 | |
| 2 | O4 Mini | 93.4 | #180 |
| 3 | LongCat-Flash-Thinking | 93.3 | #363 |
| 4 | Gemini 2.5 Pro | 92 | #152 |
| 5 | O3 | 91.6 | #124 |
| 6 | DeepSeek R1 0528 | 91.4 | #222 |
| 7 | GLM-4.5 | 91 | #275 |
| 8 | Ministral-3-14B-Reasoning-2512 | 89.8 | #540 |
| 9 | GLM 4.5 Air | 89.4 | #392 |
| 10 | Gemini 2.5 Flash | 88 | #245 |
| 11 | O3 Mini | 87.3 | #272 |
| 12 | MiniMax M1 80k | 86 | #432 |
| 13 | Ministral-3-8B-Reasoning-2512 | 86 | #721 |
| 14 | O1 Pro | 86 | #228 |
| 15 | Qwen 3 235B A22B | 85.7 | #313 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-aime-2024 · How It Works · Data refreshed daily, snapshot 2026-10-20.