# AA AIME 2025 leaderboard

Mathematics benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/math-aa-aime2025

Artificial Analysis AIME 2025. Every model the catalog carries a published AA AIME 2025 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Mathematics |
| Measure | Accuracy |
| Tasks | 30 AIME 2025 problems |
| Difficulty | Olympiad mathematics |
| Published by | [Artificial Analysis AIME 2025 Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/aime-2025) |

An independently evaluated AIME 2025 result from Artificial Analysis.

## AA AIME 2025 ranking

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 | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-5.2](https://modelchoice.dev/models/gpt-5-2) | OpenAI | 99.0 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Gemini 3 Pro](https://modelchoice.dev/models/gemini-3-pro) | Google | 95.7 | Observed 2026-11-21 · source benchlm:models |
| 3 | [GLM-4.7](https://modelchoice.dev/models/glm-4-7) | Z.AI | 95.0 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-5](https://modelchoice.dev/models/gpt-5) | OpenAI | 94.3 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-5 (high)](https://modelchoice.dev/models/gpt-5-high) | OpenAI | 94.3 | Last good 2026-11-18 · source benchlm:models |
| 6 | [GPT-5.1](https://modelchoice.dev/models/gpt-5-1) | OpenAI | 94.0 | Observed 2026-11-21 · source benchlm:models |
| 7 | [GPT-OSS 120B](https://modelchoice.dev/models/gpt-oss-120b) | OpenAI | 93.4 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Grok 4](https://modelchoice.dev/models/grok-4) | xAI | 92.7 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Claude Opus 4.5 Thinking](https://modelchoice.dev/models/claude-opus-4-5-thinking) | Anthropic | 91.3 | Observed 2026-11-21 · source benchlm:models |
| 10 | [o4-mini](https://modelchoice.dev/models/o4-mini) | OpenAI | 90.7 | Observed 2026-11-21 · source benchlm:models |
| 11 | [K-Exaone](https://modelchoice.dev/models/k-exaone) | LG AI Research | 90.3 | Observed 2026-11-21 · source benchlm:models |
| 12 | [DeepSeek V3.1 (Reasoning)](https://modelchoice.dev/models/deepseek-v3-1-reasoning) | DeepSeek | 89.7 | Last good 2026-11-18 · source benchlm:models |
| 12 | [Grok 4 Fast (Reasoning)](https://modelchoice.dev/models/grok-4-fast-reasoning) | xAI | 89.7 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
