Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Human Preference Evaluation on Arena (Phase 2)
Loading...
200
Total Battles
google/gemma-3-27b-it
182.32
186.91
191.5
196.09
May 11, 2026
Total Battles
Wins (Count)
Win Rate
Ties (Count)
Tie Rate
Losses (Count)
Loss Rate
Updated 21d ago
Evaluation Results
Method
Method
Links
Total Battles
Wins (Count)
Win Rate
Ties (Count)
Tie Rate
Losses (Count)
Loss Rate
google/gemma-3-27b-it
active parameters=27B
2026.05
200
122
61
40
20
38
19
Hebatron
active parameters=3B
2026.05
197
77
39.1
39
19.8
81
41.1
DictaLM-3.0-24B-Thinking
active parameters=23B
2026.05
183
41
22.4
21
11.5
121
66.1
Feedback
Search any
task
Search any
task