Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Mathematics benchmark
Artificial Analysis AIME 2025. Every model the catalog carries a published AA AIME 2025 value for, ranked by that value.
An independently evaluated AIME 2025 result from Artificial Analysis. 13 models in the ModelScale catalog have a published value.
Each point is a model with a published AA AIME 2025 value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
13 models with a published AA AIME 2025 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Accuracy |
|---|---|---|---|
| 1 | OpenAI | 99.0 | |
| 2 | 95.7 | ||
| 3 | Z.AI | 95.0 | |
| 4 |
What the benchmark measures and how its value is published, from the source’s own record.
AA AIME 2025 (Artificial Analysis AIME 2025) is a mathematics benchmark published by Artificial Analysis AIME 2025 Benchmark Leaderboard. An independently evaluated AIME 2025 result from Artificial Analysis. Its evaluation set is described by the source as 30 AIME 2025 problems. The source rates its difficulty as olympiad mathematics.
Each model's result is published by Artificial Analysis AIME 2025 Benchmark Leaderboard as accuracy. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 13 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| OpenAI |
| 94.3 |
| 4 | OpenAI | 94.3 |
| 6 | OpenAI | 94.0 |
| 7 | OpenAI | 93.4 |
| 8 | xAI | 92.7 |
| 9 | Anthropic | 91.3 |
| 10 | OpenAI | 90.7 |
| 11 | LG AI Research | 90.3 |
| 12 | DeepSeek | 89.7 |
| 12 | xAI | 89.7 |
Evidence key: ObservedLast good
Rows are ordered by the value Artificial Analysis AIME 2025 Benchmark Leaderboard published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.