# Mathematical Reasoning on MATH (avg@8)

Source: [Mathematical Reasoning on MATH (avg@8)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-avg-at-8>)

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

Data updated: 2026-04-15T07:52:17.991Z

Evaluation of model performance on the MATH dataset for mathematical reasoning, specifically reporting the average score at k=8 (avg@8).

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 | Average Score @8 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 14 of 14 |
| 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

- [Average Score @8](<https://www.sota2.com/research/sota/mathematical-reasoning-on-math-avg-at-8.md?metric=Average%20Score%20%408>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Score @8 | Result date | Report paper |
| --- | --- | --- | --- |
| SD-ZERO | 94 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SD-ZERO | 93.6 | Not recorded | 69df1b9a7d0172140ec7b94f |
| GRPO | 93.5 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SRT | 92.4 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SRT | 92.3 | Not recorded | 69df1b9a7d0172140ec7b94f |
| RFT | 91.9 | Not recorded | 69df1b9a7d0172140ec7b94f |
| GRPO | 91.8 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SFT | 91.4 | Not recorded | 69df1b9a7d0172140ec7b94f |
| RFT | 91.4 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SFT | 91.3 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SDFT | 91.2 | Not recorded | 69df1b9a7d0172140ec7b94f |
| SDFT | 91.1 | Not recorded | 69df1b9a7d0172140ec7b94f |
| Olmo-3-7B-Instruct | 91.1 | Not recorded | 69df1b9a7d0172140ec7b94f |
| Qwen3-4B-Instruct | 91 | Not recorded | 69df1b9a7d0172140ec7b94f |

### SD-ZERO

Average Score @8: 94. Result ID: 69df422747f34ce924c7c24b.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### SD-ZERO

Average Score @8: 93.6. Result ID: 69df422747f34ce924c7c244.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

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

### GRPO

Average Score @8: 93.5. Result ID: 69df422747f34ce924c7c241.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

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

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

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

### SRT

Average Score @8: 92.4. Result ID: 69df422747f34ce924c7c243.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

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

### SRT

Average Score @8: 92.3. Result ID: 69df422747f34ce924c7c24a.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### RFT

Average Score @8: 91.9. Result ID: 69df422747f34ce924c7c240.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa708d1e209131cea4bb20.md>)

[Implementation](<https://github.com/OFA-Sys/gsm8k-ScRel>)

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

### GRPO

Average Score @8: 91.8. Result ID: 69df422747f34ce924c7c248.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

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

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

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### SFT

Average Score @8: 91.4. Result ID: 69df422747f34ce924c7c246.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### RFT

Average Score @8: 91.4. Result ID: 69df422747f34ce924c7c247.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa708d1e209131cea4bb20.md>)

[Implementation](<https://github.com/OFA-Sys/gsm8k-ScRel>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### SFT

Average Score @8: 91.3. Result ID: 69df422747f34ce924c7c23f.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

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

### SDFT

Average Score @8: 91.2. Result ID: 69df422747f34ce924c7c249.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69799864a2b28e761d87ce97.md>)

[Implementation](<https://github.com/idanshen/Self-Distillation>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### SDFT

Average Score @8: 91.1. Result ID: 69df422747f34ce924c7c242.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69799864a2b28e761d87ce97.md>)

[Implementation](<https://github.com/idanshen/Self-Distillation>)

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

### Olmo-3-7B-Instruct

Average Score @8: 91.1. Result ID: 69df422747f34ce924c7c245.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/694238e845614693628503e0.md>)

Recorded evaluation configuration: {"Base Model":"Olmo-3-7B-Instruct"}. This record does not assert independent reproduction.

### Qwen3-4B-Instruct

Average Score @8: 91. Result ID: 69df422747f34ce924c7c23e.

[Reporting paper](<https://www.sota2.com/research/paper/69df1b9a7d0172140ec7b94f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

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

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