# Mathematical Reasoning on AIME25 (pass@1 (%)) Source: [Mathematical Reasoning on AIME25 (pass@1 (%))]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:57:30.228Z Evaluation of model performance on the AIME 2025 dataset for mathematical reasoning, measured using the Pass@1 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) | AIME25 | | Metric | Pass@1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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](): Higher is better; unit: PERCENT. ## Recorded results | Method | Pass@1 | Result date | Report paper | | --- | --- | --- | --- | | PREPO | 66.67 | Not recorded | 690988673ef7ccd74e5e5870 | | Random Selection | 60 | Not recorded | 690988673ef7ccd74e5e5870 | | GRESO | 56.67 | Not recorded | 690988673ef7ccd74e5e5870 | | Qwen3-4B | 30 | Not recorded | 690988673ef7ccd74e5e5870 | | Random Selection | 20 | Not recorded | 690988673ef7ccd74e5e5870 | | PREPO | 20 | Not recorded | 690988673ef7ccd74e5e5870 | | GRESO | 18.33 | Not recorded | 690988673ef7ccd74e5e5870 | | GRESO | 15.38 | Not recorded | 690988673ef7ccd74e5e5870 | | PREPO | 12.81 | Not recorded | 690988673ef7ccd74e5e5870 | | PREPO | 10.21 | Not recorded | 690988673ef7ccd74e5e5870 | | Random Selection | 10 | Not recorded | 690988673ef7ccd74e5e5870 | | GRESO | 9.22 | Not recorded | 690988673ef7ccd74e5e5870 | | Qwen2.5-Math-7B | 9.17 | Not recorded | 690988673ef7ccd74e5e5870 | | Random Selection | 6.98 | Not recorded | 690988673ef7ccd74e5e5870 | | Qwen2.5-Math-1.5B | 3.54 | Not recorded | 690988673ef7ccd74e5e5870 | | Qwen2.5-7B | 1.25 | Not recorded | 690988673ef7ccd74e5e5870 | ### PREPO Pass@1: 66.67. Result ID: 699afda45698665ec48f602c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B","Training Strategy":"PREPO","# Rollouts":"348K"}. This record does not assert independent reproduction. ### Random Selection Pass@1: 60. Result ID: 699afda45698665ec48f602a. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B","Training Strategy":"Random Selection","# Rollouts":"553K"}. This record does not assert independent reproduction. ### GRESO Pass@1: 56.67. Result ID: 699afda45698665ec48f602b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B","Training Strategy":"GRESO","# Rollouts":"472K"}. This record does not assert independent reproduction. ### Qwen3-4B Pass@1: 30. Result ID: 699afda45698665ec48f6029. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen3-4B","Training Strategy":"Base","# Rollouts":"–"}. This record does not assert independent reproduction. ### Random Selection Pass@1: 20. Result ID: 699afda45698665ec48f6022. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-1.5B","Training Strategy":"Random Selection","# Rollouts":"3.0M"}. This record does not assert independent reproduction. ### PREPO Pass@1: 20. Result ID: 699afda45698665ec48f6024. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-1.5B","Training Strategy":"PREPO","# Rollouts":"1.1M"}. This record does not assert independent reproduction. ### GRESO Pass@1: 18.33. Result ID: 699afda45698665ec48f6027. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-7B","Training Strategy":"GRESO","# Rollouts":"654K"}. This record does not assert independent reproduction. ### GRESO Pass@1: 15.38. Result ID: 699afda45698665ec48f6023. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-1.5B","Training Strategy":"GRESO","# Rollouts":"2.5M"}. This record does not assert independent reproduction. ### PREPO Pass@1: 12.81. Result ID: 699afda45698665ec48f6028. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-7B","Training Strategy":"PREPO","# Rollouts":"540K"}. This record does not assert independent reproduction. ### PREPO Pass@1: 10.21. Result ID: 699afda45698665ec48f6020. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training Strategy":"PREPO","# Rollouts":"304K"}. This record does not assert independent reproduction. ### Random Selection Pass@1: 10. Result ID: 699afda45698665ec48f6026. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-7B","Training Strategy":"Random Selection","# Rollouts":"905K"}. This record does not assert independent reproduction. ### GRESO Pass@1: 9.22. Result ID: 699afda45698665ec48f601f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training Strategy":"GRESO","# Rollouts":"680K"}. This record does not assert independent reproduction. ### Qwen2.5-Math-7B Pass@1: 9.17. Result ID: 699afda45698665ec48f6025. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-7B","Training Strategy":"Base","# Rollouts":"–"}. This record does not assert independent reproduction. ### Random Selection Pass@1: 6.98. Result ID: 699afda45698665ec48f601e. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training Strategy":"Random Selection","# Rollouts":"716K"}. This record does not assert independent reproduction. ### Qwen2.5-Math-1.5B Pass@1: 3.54. Result ID: 699afda45698665ec48f6021. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-Math-1.5B","Training Strategy":"Base","# Rollouts":"–"}. This record does not assert independent reproduction. ### Qwen2.5-7B Pass@1: 1.25. Result ID: 699afda45698665ec48f601d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"Qwen2.5-7B","Training Strategy":"Base","# Rollouts":"–"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.