Mathematical Reasoning on OLY
32.7AccuracyEDU-72B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EDU-72BSamples=128, Parameter Size=72B2025.03 | 32.7 | — | |
| Qwen2.5-Math-RM-72BSamples=128, Parameter Size=72B2025.03 | 30.4 | — | |
| EDU-72BSamples=64, Parameter Size=72B2025.03 | 30.2 | — | |
| Qwen2.5-Math-PRM-72BSamples=128, Parameter Size=72B2025.03 | 29.3 | — | |
| Qwen2.5-Math-RM-72BSamples=64, Parameter Size=72B2025.03 | 29.2 | — | |
| Qwen2.5-Math-PRM-7BSamples=128, Parameter Size=7B2025.03 | 28.9 | — | |
| Qwen2.5-Math-PRM-72BSamples=64, Parameter Size=72B2025.03 | 28.6 | — | |
| EDU-72BSamples=32, Parameter Size=72B2025.03 | 27.6 | — | |
| Qwen2.5-Math-RM-72BSamples=32, Parameter Size=72B2025.03 | 27.4 | — | |
| Qwen2.5-Math-PRM-72BSamples=32, Parameter Size=72B2025.03 | 27 | — | |
| Sample-EDU-72BSamples=128, Parameter Size=72B2025.03 | 27 | — | |
| Math-Shepherd-72BSamples=128, Parameter Size=72B2025.03 | 26.7 | — | |
| EDU-72BSamples=16, Parameter Size=72B2025.03 | 26.7 | — | |
| GRPOModel=OLMo-3-7B-Instruct2026.05 | 26.6 | — | |
| RRPOModel=OLMo-3-7B-Instruct2026.05 | 26.1 | — | |
| BaseModel=Qwen2.5-14B-Instruct, Note=seed-invariant2026.05 | 26 | — | |
| Qwen2.5-Math-PRM-7BSamples=64, Parameter Size=7B2025.03 | 25.8 | — | |
| Qwen2.5-Math-PRM-72BSamples=16, Parameter Size=72B2025.03 | 25.8 | — | |
| RRPOModel=Qwen2.5-14B-Instruct2026.05 | 25.7 | — | |
| Qwen2.5-Math-RM-72BSamples=16, Parameter Size=72B2025.03 | 25.5 | — | |
| Omega-72BSamples=128, Parameter Size=72B2025.03 | 25.5 | — | |
| EDU-72BSamples=8, Parameter Size=72B2025.03 | 25.5 | — | |
| SRPOModel=Qwen2.5-14B-Instruct2026.05 | 25.5 | — | |
| SCoReModel=Qwen2.5-14B-Instruct2026.05 | 25.3 | — | |
| SPO-TreeModel=Qwen2.5-14B-Instruct2026.05 | 25.1 | — | |
| GRPOModel=Qwen2.5-14B-Instruct2026.05 | 25 | — | |
| Cr-GRPOModel=Qwen2.5-14B-Instruct2026.05 | 24.9 | — | |
| SRPOModel=OLMo-3-7B-Instruct2026.05 | 24.8 | — | |
| Qwen2.5-Math-PRM-72BSamples=8, Parameter Size=72B2025.03 | 24.7 | — | |
| Math-Shepherd-72BSamples=64, Parameter Size=72B2025.03 | 24.7 | — | |
| Qwen2.5-Math-PRM-7BSamples=32, Parameter Size=7B2025.03 | 24.6 | — | |
| Omega-72BSamples=32, Parameter Size=72B2025.03 | 24.6 | — | |
| Qwen2.5-Math-RM-72BSamples=8, Parameter Size=72B2025.03 | 24.4 | — | |
| Omega-72BSamples=64, Parameter Size=72B2025.03 | 24.4 | — | |
| Sample-EDU-72BSamples=64, Parameter Size=72B2025.03 | 24.1 | — | |
| Qwen2.5-Math-PRM-7BSamples=16, Parameter Size=7B2025.03 | 23.6 | — | |
| Math-Shepherd-72BSamples=32, Parameter Size=72B2025.03 | 23.6 | — | |
| Sample-EDU-72BSamples=32, Parameter Size=72B2025.03 | 23.6 | — | |
| BaseModel=OLMo-3-7B-Instruct, Note=seed-invariant2026.05 | 23.2 | — | |
| Qwen2.5-Math-PRM-7BSamples=8, Parameter Size=7B2025.03 | 23 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=128, Parameter Size=7B2025.03 | 22.7 | — | |
| Omega-72BSamples=16, Parameter Size=72B2025.03 | 22.5 | — | |
| Math-Shepherd-72BSamples=16, Parameter Size=72B2025.03 | 22.4 | — | |
| Sample-EDU-72BSamples=16, Parameter Size=72B2025.03 | 22.4 | — | |
| EDU-72BSamples=4, Parameter Size=72B2025.03 | 22.4 | — | |
| Sample-EDU-72BSamples=8, Parameter Size=72B2025.03 | 22.2 | — | |
| Qwen2.5-Math-PRM-72BSamples=4, Parameter Size=72B2025.03 | 21.9 | — | |
| Math-Shepherd-72BSamples=8, Parameter Size=72B2025.03 | 21.9 | — | |
| Cr-GRPOModel=OLMo-3-7B-Instruct2026.05 | 21.9 | — | |
| Qwen2.5-Math-RM-72BSamples=4, Parameter Size=72B2025.03 | 21.8 | — | |
| SPO-TreeModel=OLMo-3-7B-Instruct2026.05 | 21.5 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=64, Parameter Size=7B2025.03 | 21.3 | — | |
| Omega-72BSamples=8, Parameter Size=72B2025.03 | 21.1 | — | |
| Sample-EDU-72BSamples=4, Parameter Size=72B2025.03 | 21 | — | |
| Qwen2.5-Math-PRM-7BSamples=4, Parameter Size=7B2025.03 | 20.7 | — | |
| Omega-72BSamples=4, Parameter Size=72B2025.03 | 20.7 | — | |
| Math-Shepherd-72BSamples=4, Parameter Size=72B2025.03 | 20.4 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=32, Parameter Size=7B2025.03 | 20.3 | — | |
| Sample-EDU-7BSamples=128, Parameter Size=7B2025.03 | 20.1 | — | |
| EDU-7BSamples=128, Parameter Size=7B2025.03 | 20 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=16, Parameter Size=7B2025.03 | 19.9 | — | |
| EDU-7BSamples=32, Parameter Size=7B2025.03 | 19.7 | — | |
| EDU-7BSamples=4, Parameter Size=7B2025.03 | 19.4 | — | |
| EDU-7BSamples=64, Parameter Size=7B2025.03 | 19.4 | — | |
| Qwen2.5-Math-RM-72BSamples=2, Parameter Size=72B2025.03 | 19.4 | — | |
| EDU-72BSamples=2, Parameter Size=72B2025.03 | 19.4 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=8, Parameter Size=7B2025.03 | 19.3 | — | |
| Sample-EDU-7BSamples=32, Parameter Size=7B2025.03 | 19.1 | — | |
| Sample-EDU-7BSamples=64, Parameter Size=7B2025.03 | 19.1 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=64, Parameter Size=7B2025.03 | 18.8 | — | |
| Qwen2.5-Math-PRM-72BSamples=2, Parameter Size=72B2025.03 | 18.8 | — | |
| Math-Shepherd-72BSamples=2, Parameter Size=72B2025.03 | 18.8 | — | |
| Sample-EDU-72BSamples=2, Parameter Size=72B2025.03 | 18.8 | — | |
| Sample-EDU-7BSamples=8, Parameter Size=7B2025.03 | 18.7 | — | |
| Omega-72BSamples=2, Parameter Size=72B2025.03 | 18.7 | — | |
| EDU-7BSamples=8, Parameter Size=7B2025.03 | 18.4 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=32, Parameter Size=7B2025.03 | 18.2 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=4, Parameter Size=7B2025.03 | 18.2 | — | |
| Sample-EDU-7BSamples=16, Parameter Size=7B2025.03 | 18.2 | — | |
| EDU-7BSamples=16, Parameter Size=7B2025.03 | 18.2 | — | |
| Sample-EDU-7BSamples=4, Parameter Size=7B2025.03 | 18.1 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=128, Parameter Size=7B2025.03 | 17.9 | — | |
| Qwen2.5-Math-PRM-7BSamples=2, Parameter Size=7B2025.03 | 17.9 | — | |
| Omega-7BSamples=128, Parameter Size=7B2025.03 | 17.9 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=16, Parameter Size=7B2025.03 | 17.6 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=8, Parameter Size=7B2025.03 | 17.5 | — | |
| Omega-7BSamples=16, Parameter Size=7B2025.03 | 17.5 | — | |
| Omega-7BSamples=32, Parameter Size=7B2025.03 | 17.5 | — | |
| Sample-EDU-7BSamples=2, Parameter Size=7B2025.03 | 17.5 | — | |
| SCoReModel=OLMo-3-7B-Instruct2026.05 | 17.3 | — | |
| Math-Shepherd-7BSamples=2, Parameter Size=7B2025.03 | 16.9 | — | |
| Omega-7BSamples=64, Parameter Size=7B2025.03 | 16.9 | — | |
| Math-Shepherd-7BSamples=4, Parameter Size=7B2025.03 | 16.4 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=4, Parameter Size=7B2025.03 | 16.3 | — | |
| Qwen2.5-Math-7B-PRM800KSamples=2, Parameter Size=7B2025.03 | 16 | — | |
| Omega-7BSamples=8, Parameter Size=7B2025.03 | 16 | — | |
| EDU-7BSamples=2, Parameter Size=7B2025.03 | 16 | — | |
| Math-Shepherd-Mistral-7B-PRMSamples=2, Parameter Size=7B2025.03 | 15.9 | — | |
| Math-Shepherd-7BSamples=32, Parameter Size=7B2025.03 | 15.4 | — | |
| Omega-7BSamples=4, Parameter Size=7B2025.03 | 15.3 | — |