# Mathematical Reasoning on MATH (Pass@1 Accuracy)

Source: [Mathematical Reasoning on MATH (Pass@1 Accuracy)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-pass-at-1-accuracy>)

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

Data updated: 2026-04-28T09:42:54.443Z

Evaluation of model performance on the MATH dataset for mathematical reasoning, specifically reporting the Pass@1 Accuracy metric.

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 | Mathematical Reasoning |
| Dataset (including any stated split) | MATH |
| Metric | Pass@1 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Pass@1 Accuracy](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-pass-at-1-accuracy.md?metric=Pass%401%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Pass@1 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-4 | 80 | Not recorded | 69f03f3355a3c32af89fa21d |
| IRIS | 64.6 | Not recorded | 69f03f3355a3c32af89fa21d |
| Qwen2.5 Base | 55.4 | Not recorded | 69f03f3355a3c32af89fa21d |
| DeepSeek-Math | 36.2 | Not recorded | 69f03f3355a3c32af89fa21d |
| IRIS | 29.5 | Not recorded | 69f03f3355a3c32af89fa21d |
| MetaMath | 19.8 | Not recorded | 69f03f3355a3c32af89fa21d |
| WizMath Base | 10.7 | Not recorded | 69f03f3355a3c32af89fa21d |

### GPT-4

Pass@1 Accuracy: 80. Result ID: 69f0810b47f34ce924d6d794.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### IRIS

Pass@1 Accuracy: 64.6. Result ID: 69f0810b47f34ce924d6d792.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5","Zero-shot":"true"}. This record does not assert independent reproduction.

### Qwen2.5 Base

Pass@1 Accuracy: 55.4. Result ID: 69f0810b47f34ce924d6d795.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74d81e209131cea6419f.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-Math>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### DeepSeek-Math

Pass@1 Accuracy: 36.2. Result ID: 69f0810b47f34ce924d6d798.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### IRIS

Pass@1 Accuracy: 29.5. Result ID: 69f0810b47f34ce924d6d793.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

Recorded evaluation configuration: {"Backbone":"WizMath","Zero-shot":"true"}. This record does not assert independent reproduction.

### MetaMath

Pass@1 Accuracy: 19.8. Result ID: 69f0810b47f34ce924d6d797.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73161e209131cea5a76d.md>)

[Implementation](<https://www.sota2.com/meta-math.github.io>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### WizMath Base

Pass@1 Accuracy: 10.7. Result ID: 69f0810b47f34ce924d6d796.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f3355a3c32af89fa21d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa791f1e209131cea7c3cc.md>)

[Implementation](<https://github.com/nlpxucan/WizardLM>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

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