Human Preference Evaluation on Arena (Phase 2)
200Total Battlesgoogle/gemma-3-27b-it
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| google/gemma-3-27b-itactive parameters=27B2026.05 | 200 | 122 | 61 | 40 | 20 | 38 | 19 | |
| Hebatronactive parameters=3B2026.05 | 197 | 77 | 39.1 | 39 | 19.8 | 81 | 41.1 | |
| DictaLM-3.0-24B-Thinkingactive parameters=23B2026.05 | 183 | 41 | 22.4 | 21 | 11.5 | 121 | 66.1 |