AIME 2025: scores, leaders and methodology
AIME 2025 leaderboard data from the index. Problems from the 2025 American Invitational Mathematics Examination. The top score is 91.4 and the average is 63.6.
What the data shows
Claude Opus 4.5 Pro leads at 91.4. The gap to the average is 27.8 points, and the gap to the 10th-ranked model is 2.5 points.
The median score is 67.6. Models above that line are in the upper half of the index for this benchmark.
How to read the ranking
- Use the score to shortlist, not to declare a winner.
- Check price and context for the top candidates.
- Read the benchmark methodology before quoting a score.
- Pair this benchmark with a task-specific evaluation.
Top 20 on AIME 2025
| # | Model | Provider | Score | Input / 1M | Context |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.5 Pro | Anthropic | 91.4 | $20.98 | 200,000 |
| 2 | GPT-5.2 Pro | OpenAI | 91.0 | $21.32 | 2,000,000 |
| 3 | GPT-5.1 Pro | OpenAI | 90.6 | $19.77 | 200,000 |
| 4 | GPT-5.2 | OpenAI | 90.6 | $19.71 | 2,000,000 |
| 5 | o4 Reasoning | OpenAI | 90.3 | $19.69 | 1,000,000 |
| 6 | o4 Pro | OpenAI | 89.9 | $19.37 | 1,000,000 |
| 7 | Claude Opus 4.5 | Anthropic | 89.8 | $19.42 | 400,000 |
| 8 | Gemini 3 Pro Max | Google DeepMind | 89.6 | $20.57 | 2,000,000 |
| 9 | Claude Opus 4.5 Max | Anthropic | 89.3 | $20.90 | 1,000,000 |
| 10 | Claude Opus 4.5 Reasoning | Anthropic | 88.9 | $20.91 | 2,000,000 |
| 11 | GPT-5.1 Max | OpenAI | 88.8 | $20.50 | 2,000,000 |
| 12 | Gemini 3 Pro Reasoning | Google DeepMind | 88.8 | $19.69 | 200,000 |
| 13 | Claude Sonnet 4.5 Pro | Anthropic | 88.4 | $18.73 | 1,000,000 |
| 14 | GPT-5.2 Max | OpenAI | 88.2 | $21.39 | 1,000,000 |
| 15 | Gemini 3 Pro Pro | Google DeepMind | 88.0 | $19.80 | 200,000 |
| 16 | Claude Sonnet 4.5 Turbo | Anthropic | 87.8 | $18.37 | 1,000,000 |
| 17 | Claude Sonnet 4.5 Reasoning | Anthropic | 87.3 | $19.48 | 400,000 |
| 18 | Gemini 3 Pro Vision | Google DeepMind | 87.3 | $17.59 | 200,000 |
| 19 | Grok 4 Max | xAI | 87.3 | $19.70 | 2,000,000 |
| 20 | Claude Opus 4.5 Turbo | Anthropic | 87.2 | $19.57 | 2,000,000 |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the AIME 2025: scores, leaders and methodology data show?
AIME 2025 leaderboard data from the index. Problems from the 2025 American Invitational Mathematics Examination. The top score is 91.4 and the average is 63.6.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.