Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-turn Conversation on MT-Bench (SR and Tau)

5.23SR

GTO

1.18442.23473.2854.3353Sep 26, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
5.237.01
2025.09
4.846.59
2025.09
4.15.71
2025.09
4.15.71
2025.09
4.035.64
2025.09
3.975.56
2025.09
3.95.34
2025.09
3.794.78
2025.09
3.785.4
2025.09
3.685.18
2025.09
3.495.6
2025.09
3.474.33
2025.09
3.446.15
2025.09
3.345.32
2025.09
3.275.77
2025.09
2.954.68
2025.09
2.813.67
2025.09
2.764.59
2025.09
2.714.24
2025.09
2.714.54
2025.09
2.573.25
2025.09
2.524.02
2025.09
2.494.7
2025.09
2.494.84
2025.09
2.383.93
2025.09
2.374.51
2025.09
2.374.53
2025.09
2.293.9
2025.09
2.294.37
2025.09
2.244.23
2025.09
2.143.06
2025.09
2.043.13
2025.09
1.992.48
2025.09
1.912.24
2025.09
1.732.97
2025.09
1.622.39
2025.09
1.611.67
2025.09
1.591.81
2025.09
1.581.66
2025.09
1.531.61
2025.09
1.521.61
2025.09
1.431.51
2025.09
1.341.42