# Mathematical Reasoning on DAPO-MATH Source: [Mathematical Reasoning on DAPO-MATH]() Collection index: [llms.txt]() Data updated: 2026-05-19T06:18:37.298Z Evaluation of model performance on the DAPO-MATH dataset for mathematical reasoning tasks, typically measured by metrics like Pass@1. 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) | DAPO-MATH | | 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 | | --- | --- | --- | --- | | Qwen3-4B | 80.7 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-8B | 80.7 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-4B | 79.5 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-8B | 76.8 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-4B | 57.6 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-4B | 57.6 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-7B | 54.6 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-7B | 53 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-7B | 46.2 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-1.7B | 45.2 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen3-1.7B | 45 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-7B | 44 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-1.5B | 31 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-1.5B | 28.1 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-1.5B | 26.5 | Not recorded | 6a0beebbdbc836b87d82c5bd | | Qwen2.5-Math-1.5B | 24.9 | Not recorded | 6a0beebbdbc836b87d82c5bd | ### Qwen3-4B Pass@1: 80.7. Result ID: 6a0bffbd47f34ce924f1cd09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"Base","FLOPs Budget":"8.56 × 10^19"}. This record does not assert independent reproduction. ### Qwen3-8B Pass@1: 80.7. Result ID: 6a0bffbd47f34ce924f1cd0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"LZE","FLOPs Budget":"6.30 × 10^19"}. This record does not assert independent reproduction. ### Qwen3-4B Pass@1: 79.5. Result ID: 6a0bffbd47f34ce924f1cd0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"LZE","FLOPs Budget":"5.48 × 10^19"}. This record does not assert independent reproduction. ### Qwen3-8B Pass@1: 76.8. Result ID: 6a0bffbd47f34ce924f1cd0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"Base","FLOPs Budget":"9.85 × 10^19"}. This record does not assert independent reproduction. ### Qwen3-4B Pass@1: 57.6. Result ID: 6a0bffbd47f34ce924f1cd05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"Base","FLOPs Budget":"1.18 × 10^20"}. This record does not assert independent reproduction. ### Qwen3-4B Pass@1: 57.6. Result ID: 6a0bffbd47f34ce924f1cd06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"LZE","FLOPs Budget":"7.55 × 10^19"}. This record does not assert independent reproduction. ### Qwen2.5-Math-7B Pass@1: 54.6. Result ID: 6a0bffbd47f34ce924f1cd08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"LZE","FLOPs Budget":"1.89 × 10^20"}. This record does not assert independent reproduction. ### Qwen2.5-Math-7B Pass@1: 53. Result ID: 6a0bffbd47f34ce924f1cd07. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"DAPO-MATH","Method Variant":"Base","FLOPs Budget":"2.94 × 10^20"}. This record does not assert independent reproduction. ### Qwen2.5-Math-7B Pass@1: 46.2. Result ID: 6a0bffbd47f34ce924f1cd04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"LZE","FLOPs Budget":"1.09 × 10^20"}. This record does not assert independent reproduction. ### Qwen3-1.7B Pass@1: 45.2. Result ID: 6a0bffbd47f34ce924f1cd00. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"GSM8K","Method Variant":"LZE","FLOPs Budget":"3.17 × 10^19"}. This record does not assert independent reproduction. ### Qwen3-1.7B Pass@1: 45. Result ID: 6a0bffbd47f34ce924f1ccff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"GSM8K","Method Variant":"Base","FLOPs Budget":"4.94 × 10^19"}. This record does not assert independent reproduction. ### Qwen2.5-Math-7B Pass@1: 44. Result ID: 6a0bffbd47f34ce924f1cd03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"Base","FLOPs Budget":"1.70 × 10^20"}. This record does not assert independent reproduction. ### Qwen2.5-Math-1.5B Pass@1: 31. Result ID: 6a0bffbd47f34ce924f1cd02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"LZE","FLOPs Budget":"3.73 × 10^19"}. This record does not assert independent reproduction. ### Qwen2.5-Math-1.5B Pass@1: 28.1. Result ID: 6a0bffbd47f34ce924f1cd01. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"MATH","Method Variant":"Base","FLOPs Budget":"5.82 × 10^19"}. This record does not assert independent reproduction. ### Qwen2.5-Math-1.5B Pass@1: 26.5. Result ID: 6a0bffbd47f34ce924f1ccfe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"GSM8K","Method Variant":"LZE","FLOPs Budget":"3.50 × 10^19"}. This record does not assert independent reproduction. ### Qwen2.5-Math-1.5B Pass@1: 24.9. Result ID: 6a0bffbd47f34ce924f1ccfd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Set":"GSM8K","Method Variant":"Base","FLOPs Budget":"5.47 × 10^19"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.