Question Answering on CUS-QA orig.
95.6CSQwen 3 30B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Qwen 3 30BShot=Zero2026.03 | 95.6 | 95.5 | 82.1 | |
| Llama 3.3 70BShot=Few2026.03 | 91.7 | 92.2 | 75.9 | |
| Llama 4 ScoutShot=Few2026.03 | 89.5 | 91.3 | 71.6 | |
| Qwen 3 30BShot=Few2026.03 | 87.1 | 88.5 | 68.4 | |
| Llama 4 ScoutShot=Zero2026.03 | 82.7 | 87.5 | 71.7 | |
| Llama 3.3 70BShot=Zero2026.03 | 77.7 | 77.4 | 60.5 |