Multiple-Choice Question Answering on EO and Earth Sciences MCQA Single
96.35AccuracyEVE-Instruct
Evaluation results
| Method | Links | |
|---|---|---|
| EVE-InstructSize (B)=24, Evaluation Protocol=0-shot2026.03 | 96.35 | |
| Qwen3Size (B)=30-A3, Evaluation Protocol=0-shot2026.03 | 93.02 | |
| Llama4 ScoutSize (B)=109-A17, Evaluation Protocol=0-shot2026.03 | 91.67 | |
| Gemma3Size (B)=27, Evaluation Protocol=0-shot2026.03 | 87.31 | |
| Mistral Small 3.2Size (B)=24, Evaluation Protocol=0-shot2026.03 | 83.51 |