# Mathematics on AIME 2025 (Accuracy, Thinking Token Reduction (%))

Source: [Mathematics on AIME 2025 (Accuracy, Thinking Token Reduction (%))](<https://www.sota2.com/research/sota/mathematics-on-aime-2025-accuracy-thinking-token-reduction-percent>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-23T12:18:51.906Z

Evaluation of model performance on the American Invitational Mathematics Examination (AIME) 2025 dataset, focusing on standard Accuracy and Thinking Token Reduction efficiency.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Mathematics |
| Dataset (including any stated split) | AIME 2025 |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/mathematics-on-aime-2025-accuracy-thinking-token-reduction-percent.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Thinking Token Reduction](<https://www.sota2.com/research/sota/mathematics-on-aime-2025-accuracy-thinking-token-reduction-percent.md?metric=Thinking%20Token%20Reduction>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DART | 56.7 | Not recorded | 6a3a13b9206542251d7c78ca |
| AT | 53.3 | Not recorded | 6a3a13b9206542251d7c78ca |
| NT | 53.3 | Not recorded | 6a3a13b9206542251d7c78ca |
| DART | 50 | Not recorded | 6a3a13b9206542251d7c78ca |
| AT | 40 | Not recorded | 6a3a13b9206542251d7c78ca |
| NT | 26.7 | Not recorded | 6a3a13b9206542251d7c78ca |

### DART

Accuracy: 56.7. Result ID: 6a3a7992fbc373592c38b536.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"Qwen3-8B"}. This record does not assert independent reproduction.

### AT

Accuracy: 53.3. Result ID: 6a3a7992fbc373592c38b535.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"Qwen3-8B"}. This record does not assert independent reproduction.

### NT

Accuracy: 53.3. Result ID: 6a3a7992fbc373592c38b537.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### DART

Accuracy: 50. Result ID: 6a3a7992fbc373592c38b539.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### AT

Accuracy: 40. Result ID: 6a3a7992fbc373592c38b538.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"DeepSeek-V3.2"}. This record does not assert independent reproduction.

### NT

Accuracy: 26.7. Result ID: 6a3a7992fbc373592c38b534.

[Reporting paper](<https://www.sota2.com/research/paper/6a3a13b9206542251d7c78ca.md>)

Recorded evaluation configuration: {"Model":"Qwen3-8B"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.