# Mathematical Reasoning on MATH500 (Pass@1 @16, Token Cost, AE Score) Source: [Mathematical Reasoning on MATH500 (Pass@1 @16, Token Cost, AE Score)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:57:32.017Z Evaluation of model performance on the MATH500 dataset for mathematical reasoning, reporting Pass@1 score averaged over 16 samples, average token usage, and AE Score. 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) | MATH500 | | Metric | Pass@1 (Avg @16) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | 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 (Avg @16)](): Higher is better; unit: PERCENT. - [Avg. Token Count](): Lower is better; unit: SCALAR. - [AE Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Pass@1 (Avg @16) | Result date | Report paper | | --- | --- | --- | --- | | DLER-R1-7B-Research | 91.8 | Not recorded | 6981817ba8b01e76cc8e2572 | | Laser-DE-L4096-7B | 91.5 | Not recorded | 6981817ba8b01e76cc8e2572 | | APR-7B (Ours) | 90.3 | Not recorded | 6981817ba8b01e76cc8e2572 | | L1-Qwen-7B-Max | 90.2 | Not recorded | 6981817ba8b01e76cc8e2572 | | TrainingEfficient DS-7B | 89.1 | Not recorded | 6981817ba8b01e76cc8e2572 | | AdaptThink-7B-delta0.05 | 88.2 | Not recorded | 6981817ba8b01e76cc8e2572 | | DLER-R1-1.5B-Research | 86.9 | Not recorded | 6981817ba8b01e76cc8e2572 | | Original Model | 86.7 | Not recorded | 6981817ba8b01e76cc8e2572 | | Original Model | 85.1 | Not recorded | 6981817ba8b01e76cc8e2572 | | Laser-L8192-1.5B | 84.9 | Not recorded | 6981817ba8b01e76cc8e2572 | | L1-Qwen-1.5B-Max | 84.8 | Not recorded | 6981817ba8b01e76cc8e2572 | | APR-1.5B (Ours) | 84.7 | Not recorded | 6981817ba8b01e76cc8e2572 | | DS-1.5B-thinkprune-iter2k | 83.1 | Not recorded | 6981817ba8b01e76cc8e2572 | | Laser-DE-L4096-1.5B | 82.8 | Not recorded | 6981817ba8b01e76cc8e2572 | | SB DS7B alpha 2 | 82.6 | Not recorded | 6981817ba8b01e76cc8e2572 | | TrainingEfficient DS-1.5B | 81.8 | Not recorded | 6981817ba8b01e76cc8e2572 | | AdaptThink-1.5B-delta0.05 | 80.8 | Not recorded | 6981817ba8b01e76cc8e2572 | ### DLER-R1-7B-Research Pass@1 (Avg @16): 91.8. Result ID: 699659a05698665ec49f1b63. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### Laser-DE-L4096-7B Pass@1 (Avg @16): 91.5. Result ID: 699659a05698665ec49f1b62. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### APR-7B (Ours) Pass@1 (Avg @16): 90.3. Result ID: 699659a05698665ec49f1b64. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### L1-Qwen-7B-Max Pass@1 (Avg @16): 90.2. Result ID: 699659a05698665ec49f1b5f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### TrainingEfficient DS-7B Pass@1 (Avg @16): 89.1. Result ID: 699659a05698665ec49f1b60. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### AdaptThink-7B-delta0.05 Pass@1 (Avg @16): 88.2. Result ID: 699659a05698665ec49f1b5e. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### DLER-R1-1.5B-Research Pass@1 (Avg @16): 86.9. Result ID: 699659a05698665ec49f1b5b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### Original Model Pass@1 (Avg @16): 86.7. Result ID: 699659a05698665ec49f1b5d. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### Original Model Pass@1 (Avg @16): 85.1. Result ID: 699659a05698665ec49f1b54. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### Laser-L8192-1.5B Pass@1 (Avg @16): 84.9. Result ID: 699659a05698665ec49f1b59. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### L1-Qwen-1.5B-Max Pass@1 (Avg @16): 84.8. Result ID: 699659a05698665ec49f1b56. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### APR-1.5B (Ours) Pass@1 (Avg @16): 84.7. Result ID: 699659a05698665ec49f1b5c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### DS-1.5B-thinkprune-iter2k Pass@1 (Avg @16): 83.1. Result ID: 699659a05698665ec49f1b58. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### Laser-DE-L4096-1.5B Pass@1 (Avg @16): 82.8. Result ID: 699659a05698665ec49f1b5a. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### SB DS7B alpha 2 Pass@1 (Avg @16): 82.6. Result ID: 699659a05698665ec49f1b61. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-7B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### TrainingEfficient DS-1.5B Pass@1 (Avg @16): 81.8. Result ID: 699659a05698665ec49f1b57. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. ### AdaptThink-1.5B-delta0.05 Pass@1 (Avg @16): 80.8. Result ID: 699659a05698665ec49f1b55. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"DeepSeek-R1-Distill-Qwen-1.5B","Sampling Temperature":0.6,"Max Response Length":8192,"Rollout samples":16}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.