# Mathematical Reasoning on OlympiadBench (Last/Best Scores)

Source: [Mathematical Reasoning on OlympiadBench (Last/Best Scores)](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiadbench-last-best-scores>)

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

Data updated: 2026-02-26T05:48:26.037Z

Evaluation of model performance on mathematical reasoning problems using the OlympiadBench dataset, reporting the final ('Last') and highest ('Best') scores achieved.

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) | OlympiadBench |
| Metric | Last Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Last Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiadbench-last-best-scores.md?metric=Last%20Score>): Higher is better; unit: PERCENT.

- [Best Score](<https://www.sota2.com/research/sota/mathematical-reasoning-on-olympiadbench-last-best-scores.md?metric=Best%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Last Score | Result date | Report paper |
| --- | --- | --- | --- |
| DAPO | 51.63 | Not recorded | 68fa7ba21e209131cea897d4 |
| RPG-REINFORCE-UFKL | 50.74 | Not recorded | 68fa7ba21e209131cea897d4 |
| REINFORCE++-Baseline | 49.26 | Not recorded | 68fa7ba21e209131cea897d4 |
| GRPO | 49.26 | Not recorded | 68fa7ba21e209131cea897d4 |
| RPG-URKL | 48.37 | Not recorded | 68fa7ba21e209131cea897d4 |
| REINFORCE++ | 47.78 | Not recorded | 68fa7ba21e209131cea897d4 |
| RPG-UFKL | 46.88 | Not recorded | 68fa7ba21e209131cea897d4 |
| RPG-REINFORCE-URKL | 46.74 | Not recorded | 68fa7ba21e209131cea897d4 |

### DAPO

Last Score: 51.63. Result ID: 699af90a5698665ec48a8368.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### RPG-REINFORCE-UFKL

Last Score: 50.74. Result ID: 699af90a5698665ec48a836b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### REINFORCE++-Baseline

Last Score: 49.26. Result ID: 699af90a5698665ec48a8365.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a631e209131cea82d06.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### GRPO

Last Score: 49.26. Result ID: 699af90a5698665ec48a8367.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

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

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

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### RPG-URKL

Last Score: 48.37. Result ID: 699af90a5698665ec48a836a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### REINFORCE++

Last Score: 47.78. Result ID: 699af90a5698665ec48a8366.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a631e209131cea82d06.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### RPG-UFKL

Last Score: 46.88. Result ID: 699af90a5698665ec48a8369.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

### RPG-REINFORCE-URKL

Last Score: 46.74. Result ID: 699af90a5698665ec48a836c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba21e209131cea897d4.md>)

Recorded evaluation configuration: {"Context Length":"4k","Base Model":"Qwen-2.5-7B-Instruct"}. This record does not assert independent reproduction.

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