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

Instruction Following on MT-Bench zh

6.83Score

Qwen2.5-14B-SFT-TaP

4.66685.22845.796.3516Jun 30, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.06
6.83
2025.06
6.58
2025.06
6.51
2025.06
6.46
2025.06
6.43
2025.06
6.41
2025.06
6.38
2025.06
6.36
2025.06
6.35
2025.06
6.32
2025.06
6.31
2025.06
6.25
2025.06
6.21
2025.06
6.19
2025.06
6.08
2025.06
6
2025.06
5.99
2025.06
5.86
2025.06
5.86
2025.06
5.81
2025.06
5.8
2025.06
5.79
2025.06
5.78
2025.06
5.78
2025.06
5.78
2025.06
5.76
2025.06
5.71
2025.06
5.68
2025.06
5.63
2025.06
5.52
2025.06
5.5
2025.06
5.49
2025.06
5.46
2025.06
5.45
2025.06
5.39
2025.06
5.28
2025.06
5.27
2025.06
5.26
2025.06
5.25
2025.06
5.25
2025.06
5.23
2025.06
5.21
2025.06
5.21
2025.06
5.2
2025.06
5.19
2025.06
5.18
2025.06
5.14
2025.06
5.14
2025.06
5.11
2025.06
5.04
2025.06
5.04
2025.06
5.03
2025.06
4.99
2025.06
4.94
2025.06
4.93
2025.06
4.93
2025.06
4.89
2025.06
4.87
2025.06
4.85
2025.06
4.75