Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Response Quality Evaluation on MT-Bench

8.71Average Response Quality

No defense

5.44446.29227.147.9878Feb 26, 2024May 11, 2024Jul 25, 2024Oct 8, 2024Dec 22, 2024Mar 7, 2025May 22, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2024.02
8.71
2024.02
8.6
2024.02
8.58
2024.02
8.43
2024.02
7.36
2024.02
7.35
2024.02
7.3
2024.02
7.26
2024.02
7.23
2024.02
6.8
2025.05
6.78
2025.05
6.76
2024.02
6.74
2024.02
6.69
2024.02
6.34
2024.02
5.89
2024.02
5.81
2025.05
5.64
2025.05
5.57