Commonsense Reasoning on ARC-e, ARC-c, PIQA, Winogrande, HellaSwag
83.6ARC-e Accuracy
Qwen3
Updated 25d ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.06 | 83.6 | 56.6 | 76.8 | 67.7 | 75 | 71.9 | |
2026.06 | 82.5 | 57.4 | 77.2 | 69.9 | 72.7 | 71.9 | |
2026.06 | 81.9 | 57.8 | 77.2 | 71.1 | 73.1 | 72.2 | |
2026.06 | 81.4 | 56.5 | 77.4 | 70.1 | 73.7 | 71.8 | |
2026.06 | 80.9 | 56.1 | 79.9 | 72.2 | 75 | 72.8 | |
2026.06 | 79.5 | 56.1 | 79.5 | 70.9 | 77.7 | 72.7 | |
2026.06 | 79.3 | 51.9 | 77.9 | 66.5 | 73.1 | 69.8 | |
2025.12 | 77.69 | 53.58 | 80.63 | 72.69 | 79.16 | 72.75 | |
2025.12 | 76.05 | 50 | 76.55 | 72.93 | 73.37 | 69.78 | |
2026.06 | 75.6 | 47.9 | 76.4 | 61.6 | 68.8 | 66 | |
2026.06 | 73.7 | 42.3 | 72.9 | 61.6 | 60.3 | 62.1 | |
2026.06 | 73.2 | 42.6 | 73 | 62.8 | 60.1 | 62.4 | |
2026.06 | 73.1 | 43.3 | 73 | 61.8 | 60.1 | 62.3 | |
2026.06 | 72.7 | 42.8 | 72.4 | 58.8 | 59.6 | 61.3 | |
2026.06 | 72.5 | 41.9 | 72.6 | 63.7 | 60 | 62.1 | |
2026.06 | 72.4 | 43.5 | 72.5 | 61 | 60.4 | 62 | |
2025.12 | 70.37 | 46.33 | 74.16 | 71.82 | 70.12 | 66.56 | |
2025.12 | 67.05 | 41.13 | 75.52 | 69.61 | 66.49 | 63.96 | |
2026.06 | 62.9 | 32.5 | 67.1 | 55.7 | 46.5 | 52.9 | |
2026.06 | 62.9 | 30.6 | 67 | 55.1 | 46 | 52.3 | |
2026.06 | 61.9 | 33.1 | 67.8 | 53.9 | 46.4 | 52.6 | |
2026.06 | 61.2 | 33.1 | 66.8 | 56.6 | 44.7 | 52.1 | |
2026.06 | 60.6 | 34.1 | 67.6 | 55.7 | 47.3 | 53.1 | |
2025.12 | 44.44 | 29.27 | 57.56 | 58.48 | 41.08 | 46.17 | |
2025.12 | 38.13 | 31.4 | 60.94 | 54.22 | 31.52 | 43.24 |