GSM8k gsm8k Leaderboard
Auto-discovered from papers reporting GSM8k (Accuracy). Β· Metric: Accuracy (higher is better) Β· π’ Updated 20h ago
| # | Model | Accuracy | Paper |
|---|---|---|---|
| 1 | QwQ-32B-Preview | 95.22 | link |
| 2 | GPT-OSS (120B) | 94.25 | link |
| 3 | DeepSeek-r1-Dist. | 91.28 | link |
| 4 | Skywork-o1 | 91.28 | link |
| 5 | Ours (+SFT) | 88.20 | link |
| 6 | STAIR | 87.60 | link |
| 7 | Self-Critique | 86.20 | link |
| 8 | Shallow-Align | 85.80 | link |
| 9 | Original | 84.50 | link |
| 10 | DPO | 84.15 | link |
| 11 | DeepSeek-R1 (70B) | 82.95 | link |
| 12 | Gemma4 (31B) | 81.35 | link |
| 13 | LLaMA-o1 | 79.38 | link |
| 14 | SFT | 72.02 | link |