Mathematical Reasoning on AIME 2024 (Pass@1, Pass@128, Weighted Accuracy)
83.33Weighted AccuracyOTV
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| OTVBackbone=QWEN3-4B2026.03 | 83.33 | — | — | |
| GenRMBackbone=QWEN3-4B2026.03 | 79.11 | — | — | |
| Qwen2.5-PRM-7BBackbone=QWEN3-4B2026.03 | 78.18 | — | — | |
| DeepConfBackbone=QWEN3-4B2026.03 | 77.76 | — | — | |
| VersaPRM-8BBackbone=QWEN3-4B2026.03 | 75.52 | — | — | |
| MAJ@128Backbone=QWEN3-4B2026.03 | 75.42 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=QWEN3-4B2026.03 | 75.16 | — | — | |
| Math-Shepherd-7BBackbone=QWEN3-4B2026.03 | 74.27 | — | — | |
| OTVBackbone=DAPO-QWEN-32B2026.03 | 70.83 | — | — | |
| AceMath-RM-7BBackbone=QWEN3-4B2026.03 | 67.66 | — | — | |
| Qwen2.5-PRM-7BBackbone=DAPO-QWEN-32B2026.03 | 67.45 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=DAPO-QWEN-32B2026.03 | 67.03 | — | — | |
| VersaPRM-8BBackbone=DAPO-QWEN-32B2026.03 | 66.98 | — | — | |
| DeepConfBackbone=DAPO-QWEN-32B2026.03 | 66.77 | — | — | |
| MAJ@128Backbone=DAPO-QWEN-32B2026.03 | 66.72 | — | — | |
| GenRMBackbone=DAPO-QWEN-32B2026.03 | 66.72 | — | — | |
| Math-Shepherd-7BBackbone=DAPO-QWEN-32B2026.03 | 65.73 | — | — | |
| AceMath-RM-7BBackbone=DAPO-QWEN-32B2026.03 | 61.15 | — | — | |
| OTVBackbone=QWEN3-8B2026.03 | 46.56 | — | — | |
| DeepConfBackbone=QWEN3-8B2026.03 | 45.73 | — | — | |
| Qwen2.5-PRM-7BBackbone=QWEN3-8B2026.03 | 45.1 | — | — | |
| AceMath-RM-7BBackbone=QWEN3-8B2026.03 | 44.9 | — | — | |
| VersaPRM-8BBackbone=QWEN3-8B2026.03 | 44.48 | — | — | |
| GenRMBackbone=QWEN3-8B2026.03 | 44.38 | — | — | |
| MAJ@128Backbone=QWEN3-8B2026.03 | 44.22 | — | — | |
| Math-Shepherd-7BBackbone=QWEN3-8B2026.03 | 44.22 | — | — | |
| Qwen2.5-PRM800K-7BBackbone=QWEN3-8B2026.03 | 44.22 | — | — | |
| Pass@1Backbone=QWEN3-4B2026.03 | — | — | 60.29 | |
| Pass@1Backbone=QWEN3-8B2026.03 | — | — | 26.29 | |
| Pass@1Backbone=DAPO-QWEN-32B2026.03 | — | — | 51.77 | |
| Pass@128Backbone=QWEN3-4B2026.03 | — | 91.46 | — | |
| Pass@128Backbone=QWEN3-8B2026.03 | — | 69.58 | — | |
| Pass@128Backbone=DAPO-QWEN-32B2026.03 | — | 83.75 | — |