Math reasoning on OlympiadBench (Pass@1)
55.6Pass@1OpenThinker3-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenThinker3-7B2026.03 | 55.6 | |
| AAPO-1.5BZero-shot=true, Number of parameters=1.5B2025.05 | 53.3 | |
| GPG-1.5BZero-shot=true, Number of parameters=1.5B2025.05 | 52.4 | |
| GRPO-1.5BZero-shot=true, Number of parameters=1.5B2025.05 | 51.3 | |
| Sampled-token OPDmasking=true2026.03 | 44.7 | |
| AAPO-7BZero-shot=true, Number of parameters=7B2025.05 | 44.3 | |
| teacher top-K local support matchingmasking=false2026.03 | 43.9 | |
| teacher top-K local support matchingmasking=true2026.03 | 43.9 | |
| Sampled-token OPDmasking=false2026.03 | 43.1 | |
| GPG-7BZero-shot=true, Number of parameters=7B2025.05 | 42.4 | |
| Oat-Zero-7BZero-shot=true, Number of parameters=7B2025.05 | 41 | |
| SimpleRL-Zero-7BZero-shot=true, Number of parameters=7B2025.05 | 40.4 | |
| Eurus-2-7B-PRIMEZero-shot=true, Number of parameters=7B2025.05 | 40.3 | |
| Qwen2.5-7B-It2026.03 | 32.9 |