Open-Ended Question Answering on Satcom Open-Ended
83Judge ScoreQwen3
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3Size (B)=30-A3, shot=0-shot2026.03 | 83 | |
| Gemma3Size (B)=27, shot=0-shot2026.03 | 81.25 | |
| EVE-InstructSize (B)=24, shot=0-shot2026.03 | 81.25 | |
| Mistral Small 3.2Size (B)=24, shot=0-shot2026.03 | 76.5 | |
| Llama4 ScoutSize (B)=109-A17, shot=0-shot2026.03 | 75 |