| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| AIME 2025 | SELF-THOUGHT | Acc@t180 | 24 | 4mo ago | |
| AIME 2024 | SELF-THOUGHT | Acc@t186.67 | 24 | 4mo ago | |
| Competition Math Average | ParaGator-Zero-4B | Pass@134 | 20 | 4mo ago | |
| Olympiad | ParaGator-Zero-4B | Pass@154.89 | 20 | 4mo ago | |
| AIME 2025 | Pass@127.37 | 20 | 4mo ago | ||
| BrumoMath 2025 | ParaGator-Zero-4B | Pass@136.25 | 20 | 4mo ago | |
| AIME 2025 | PAPO | Accuracy (avg@4)30.7 | 12 | 3mo ago | |
| AIME 2024 | PAPO | Accuracy (avg@4)34.5 | 12 | 3mo ago | |
| OlympiadBench | PAPO | Accuracy (avg@4)61.1 | 12 | 3mo ago | |
| MATH 5,000 problems (test) | TRI (SFT + DPO) | L1 Accuracy97.6 | 9 | 1mo ago | |
| Beyond AIME | REASONING ARENA-Live | Pass Ratio36.4 | 6 | 1mo ago | |
| AIME 2026 | REASONING ARENA-Live | Pass Ratio59 | 6 | 1mo ago | |
| AIME 2025 | REASONING ARENA | Pass Ratio54.8 | 6 | 1mo ago | |
| AIME 2024 | REASONING ARENA | Pass Ratio66.7 | 6 | 1mo ago | |
| IMO (test) | PoT | Efficiency Ratio0.52 | 4 | 4mo ago |