# Mathematics on AIME 24 (AVG@32, Pass@32) Source: [Mathematics on AIME 24 (AVG@32, Pass@32)]() Collection index: [llms.txt]() Data updated: 2026-07-07T08:12:03.650Z Evaluation of model performance on the American Invitational Mathematics Examination (AIME) 2024 dataset, measured using the AVG@32 and Pass@32 metrics. 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 | Math | | Dataset (including any stated split) | AIME 24 | | Metric | AVG@32 | | 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 - [AVG@32](): Higher is better; unit: PERCENT. - [Pass@32](): Higher is better; unit: PERCENT. ## Recorded results | Method | AVG@32 | Result date | Report paper | | --- | --- | --- | --- | | Projected | 74.38 | Not recorded | 6a4c87e085dba69fef8b238b | | OR-Zero | 74.27 | Not recorded | 6a4c87e085dba69fef8b238b | | Best Single-Domain Expert | 74.27 | Not recorded | 6a4c87e085dba69fef8b238b | | TIES Merging | 73.54 | Not recorded | 6a4c87e085dba69fef8b238b | | DARE + TIES | 73.06 | Not recorded | 6a4c87e085dba69fef8b238b | | DeepCoder | 71.04 | Not recorded | 6a4c87e085dba69fef8b238b | | Task Arithmetic | 71.04 | Not recorded | 6a4c87e085dba69fef8b238b | | Distill-Qwen-14B | 67.71 | Not recorded | 6a4c87e085dba69fef8b238b | | Projected | 43.44 | Not recorded | 6a4c87e085dba69fef8b238b | | TIES Merging | 42.08 | Not recorded | 6a4c87e085dba69fef8b238b | | DeepScaleR | 40.31 | Not recorded | 6a4c87e085dba69fef8b238b | | Best Single-Domain Expert | 40.31 | Not recorded | 6a4c87e085dba69fef8b238b | | DARE + TIES | 40.05 | Not recorded | 6a4c87e085dba69fef8b238b | | Archer-Code | 39.48 | Not recorded | 6a4c87e085dba69fef8b238b | | Task Arithmetic | 36.25 | Not recorded | 6a4c87e085dba69fef8b238b | | Distill-Qwen-1.5B | 31.67 | Not recorded | 6a4c87e085dba69fef8b238b | ### Projected AVG@32: 74.38. Result ID: 6a4cb4b860b88bb339c83f7e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### OR-Zero AVG@32: 74.27. Result ID: 6a4cb4b860b88bb339c83f78. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Math Expert"}. This record does not assert independent reproduction. ### Best Single-Domain Expert AVG@32: 74.27. Result ID: 6a4cb4b860b88bb339c83f7a. [Reporting paper]() Recorded evaluation configuration: {"Scale":"14B"}. This record does not assert independent reproduction. ### TIES Merging AVG@32: 73.54. Result ID: 6a4cb4b860b88bb339c83f7c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### DARE + TIES AVG@32: 73.06. Result ID: 6a4cb4b860b88bb339c83f7d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### DeepCoder AVG@32: 71.04. Result ID: 6a4cb4b860b88bb339c83f79. [Reporting paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Code Expert"}. This record does not assert independent reproduction. ### Task Arithmetic AVG@32: 71.04. Result ID: 6a4cb4b860b88bb339c83f7b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### Distill-Qwen-14B AVG@32: 67.71. Result ID: 6a4cb4b860b88bb339c83f77. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"14B","Category":"Single-Domain Reference"}. This record does not assert independent reproduction. ### Projected AVG@32: 43.44. Result ID: 6a4cb4b860b88bb339c83f76. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### TIES Merging AVG@32: 42.08. Result ID: 6a4cb4b860b88bb339c83f74. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### DeepScaleR AVG@32: 40.31. Result ID: 6a4cb4b860b88bb339c83f70. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Math Expert"}. This record does not assert independent reproduction. ### Best Single-Domain Expert AVG@32: 40.31. Result ID: 6a4cb4b860b88bb339c83f72. [Reporting paper]() Recorded evaluation configuration: {"Scale":"1.5B"}. This record does not assert independent reproduction. ### DARE + TIES AVG@32: 40.05. Result ID: 6a4cb4b860b88bb339c83f75. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### Archer-Code AVG@32: 39.48. Result ID: 6a4cb4b860b88bb339c83f71. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Code Expert"}. This record does not assert independent reproduction. ### Task Arithmetic AVG@32: 36.25. Result ID: 6a4cb4b860b88bb339c83f73. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Cross-Domain Merging"}. This record does not assert independent reproduction. ### Distill-Qwen-1.5B AVG@32: 31.67. Result ID: 6a4cb4b860b88bb339c83f6f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Scale":"1.5B","Category":"Single-Domain Reference"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.