# Mathematical Reasoning on Mathematical Reasoning Aggregate Average of 6 Benchmarks (Mean@8)

Source: [Mathematical Reasoning on Mathematical Reasoning Aggregate Average of 6 Benchmarks (Mean@8)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathematical-reasoning-aggregate-average-of-6-benchmarks-mean-at-8>)

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

Data updated: 2026-05-19T09:52:02.455Z

Evaluation of model performance on the aggregate average of 6 specific mathematical reasoning benchmarks, reporting the Mean@8 score.

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) | Mathematical Reasoning Aggregate Average of 6 Benchmarks |
| Metric | Mean@8 Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Mean@8 Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-mathematical-reasoning-aggregate-average-of-6-benchmarks-mean-at-8.md?metric=Mean%408%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Mean@8 Score | Result date | Report paper |
| --- | --- | --- | --- |
| DISA | 65.5 | Not recorded | 6a0bee98dbc836b87d82c430 |
| FlowRL | 64.8 | Not recorded | 6a0bee98dbc836b87d82c430 |
| GSPO | 62.7 | Not recorded | 6a0bee98dbc836b87d82c430 |
| GRPO | 49.4 | Not recorded | 6a0bee98dbc836b87d82c430 |
| DISA | 33.5 | Not recorded | 6a0bee98dbc836b87d82c430 |
| FlowRL | 33.2 | Not recorded | 6a0bee98dbc836b87d82c430 |
| GRPO | 32.8 | Not recorded | 6a0bee98dbc836b87d82c430 |
| Vanilla | 29.7 | Not recorded | 6a0bee98dbc836b87d82c430 |
| GSPO | 29.3 | Not recorded | 6a0bee98dbc836b87d82c430 |
| Vanilla | 23.4 | Not recorded | 6a0bee98dbc836b87d82c430 |

### DISA

Mean@8 Score: 65.5. Result ID: 6a0c314afbc373592c02b6ee.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B-Base"}. This record does not assert independent reproduction.

### FlowRL

Mean@8 Score: 64.8. Result ID: 6a0c314afbc373592c02b6ed.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc71e209131cea8a674.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B-Base"}. This record does not assert independent reproduction.

### GSPO

Mean@8 Score: 62.7. Result ID: 6a0c314afbc373592c02b6ec.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B-Base"}. This record does not assert independent reproduction.

### GRPO

Mean@8 Score: 49.4. Result ID: 6a0c314afbc373592c02b6eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

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

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

Recorded evaluation configuration: {"Backbone":"Qwen3-4B-Base"}. This record does not assert independent reproduction.

### DISA

Mean@8 Score: 33.5. Result ID: 6a0c314afbc373592c02b6e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

### FlowRL

Mean@8 Score: 33.2. Result ID: 6a0c314afbc373592c02b6e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bc71e209131cea8a674.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

### GRPO

Mean@8 Score: 32.8. Result ID: 6a0c314afbc373592c02b6e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

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

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

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

### Vanilla

Mean@8 Score: 29.7. Result ID: 6a0c314afbc373592c02b6ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B-Base"}. This record does not assert independent reproduction.

### GSPO

Mean@8 Score: 29.3. Result ID: 6a0c314afbc373592c02b6e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

### Vanilla

Mean@8 Score: 23.4. Result ID: 6a0c314afbc373592c02b6e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a0bee98dbc836b87d82c430.md>)

Recorded evaluation configuration: {"Backbone":"Qwen2.5-7B"}. This record does not assert independent reproduction.

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