# Mathematical Reasoning on MATH 500 (Accuracy and Thinking Token Usage) Source: [Mathematical Reasoning on MATH 500 (Accuracy and Thinking Token Usage)]() Collection index: [llms.txt]() Data updated: 2026-06-23T12:18:51.767Z Evaluation of model performance on mathematical reasoning problems using the MATH 500 dataset, reporting standard accuracy alongside the percentage of thinking tokens utilized during inference. 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 500 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. - [Thinking Token Usage](): Lower is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DART | 88.5 | Not recorded | 6a3a13b9206542251d7c78ca | | DART | 88.2 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 87.6 | Not recorded | 6a3a13b9206542251d7c78ca | | DART | 87.6 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 86.2 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 85.6 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 82.2 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 81.2 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 76.6 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 70.5 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 66.5 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 65.5 | Not recorded | 6a3a13b9206542251d7c78ca | | DART | 65.5 | Not recorded | 6a3a13b9206542251d7c78ca | | DART | 58 | Not recorded | 6a3a13b9206542251d7c78ca | | AT | 56 | Not recorded | 6a3a13b9206542251d7c78ca | | DART | 54 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 52.5 | Not recorded | 6a3a13b9206542251d7c78ca | | NT | 38.5 | Not recorded | 6a3a13b9206542251d7c78ca | ### DART Accuracy: 88.5. Result ID: 6a3a7992fbc373592c38b500. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"32B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### DART Accuracy: 88.2. Result ID: 6a3a7992fbc373592c38b4fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"8B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### AT Accuracy: 87.6. Result ID: 6a3a7992fbc373592c38b4fc. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"14B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### DART Accuracy: 87.6. Result ID: 6a3a7992fbc373592c38b4fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"14B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### AT Accuracy: 86.2. Result ID: 6a3a7992fbc373592c38b4ff. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"32B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### AT Accuracy: 85.6. Result ID: 6a3a7992fbc373592c38b4f9. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"8B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### NT Accuracy: 82.2. Result ID: 6a3a7992fbc373592c38b4fe. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"32B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. ### NT Accuracy: 81.2. Result ID: 6a3a7992fbc373592c38b4fb. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"14B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. ### NT Accuracy: 76.6. Result ID: 6a3a7992fbc373592c38b4f8. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"8B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. ### AT Accuracy: 70.5. Result ID: 6a3a7992fbc373592c38b4f6. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"4B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### NT Accuracy: 66.5. Result ID: 6a3a7992fbc373592c38b4f5. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"4B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. ### AT Accuracy: 65.5. Result ID: 6a3a7992fbc373592c38b4f3. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"1.7B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### DART Accuracy: 65.5. Result ID: 6a3a7992fbc373592c38b4f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"4B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### DART Accuracy: 58. Result ID: 6a3a7992fbc373592c38b4f4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"1.7B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### AT Accuracy: 56. Result ID: 6a3a7992fbc373592c38b4f0. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"0.6B","Thinking Strategy":"AT"}. This record does not assert independent reproduction. ### DART Accuracy: 54. Result ID: 6a3a7992fbc373592c38b4f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Scale":"0.6B","Thinking Strategy":"DART"}. This record does not assert independent reproduction. ### NT Accuracy: 52.5. Result ID: 6a3a7992fbc373592c38b4f2. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"1.7B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. ### NT Accuracy: 38.5. Result ID: 6a3a7992fbc373592c38b4ef. [Reporting paper]() Recorded evaluation configuration: {"Model Scale":"0.6B","Thinking Strategy":"NT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.