Avg avg Leaderboard
Auto-discovered from papers reporting Avg (Accuracy). Β· Metric: Accuracy (higher is better) Β· π’ Updated 1h ago
| # | Model | Accuracy | Paper |
|---|---|---|---|
| 1 | open-ended | 56.87 | link |
| 2 | mcq + open-ended | 55.66 | link |
| 3 | Nemotron-CrossThink (B_nd) | 55.66 | link |
| 4 | orz | 55.20 | link |
| 5 | Short | 54.50 | link |
| 6 | Long | 53.30 | link |
| 7 | Qwen2.5-Math +P-GRPO (Math+RM) | 51.50 | link |
| 8 | Eurus-2-PRIME | 48.20 | link |
| 9 | Qwen2.5-Math +GRPO (Math) | 48.00 | link |
| 10 | β³ | 44.75 | link |
| 11 | Qwen2.5-Math-Inst | 43.40 | link |
| 12 | GPT-4o | 40.80 | link |
| 13 | Llama-3.1-Inst (70B) | 38.90 | link |
| 14 | Qwen2.5-Math | 24.50 | link |