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

Multi-turn Conversation on MT-Bench (Speedup and Average Acceptance Length)

4.64Speedup

Eagle3+LTD

0.85441.83722.823.8028Mar 2, 2026Mar 15, 2026Mar 28, 2026Apr 10, 2026Apr 23, 2026May 6, 2026May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.03
4.648.24
2026.03
4.537.99
2026.03
4.487.22
2026.03
4.487.08
2026.03
4.337.22
2026.03
4.256.78
2026.03
4.156.69
2026.03
4.156.85
2026.03
4.056.32
2026.03
4.027.74
2026.03
3.966.74
2026.03
3.836.75
2026.03
3.826.09
2026.03
3.86.39
2026.03
3.795.86
2026.03
3.76.39
2026.03
3.76.11
2026.03
3.696.42
2026.03
3.675.41
2026.03
3.675.82
2026.03
3.656.21
2026.05
3.373.96
2026.05
3.254.22
2026.05
3.163.89
2026.05
2.853.57
2026.05
2.783.26
2026.05
2.783.26
2026.05
2.763.37
2026.05
2.743.39
2026.05
2.683.79
2026.05
2.683.79
2026.05
2.673.15
2026.05
2.643.05
2026.05
2.6-
2026.05
2.59-
2026.05
2.584.16
2026.05
2.573.1
2026.05
2.543.6
2026.05
2.534.29
2026.05
2.47-
2026.03
2.443.34
2026.03
2.423.17
2026.03
2.343.4
2026.05
2.273.03
2026.05
2.23-
2026.05
1.893.04
2026.03
1.763.3
2026.05
1.76-
2026.05
1.74-
2026.05
1.721.97
2026.05
1.72-
2026.05
1.712.05
2026.05
1.66-
2026.05
1.66-
2026.05
1.66-
2026.05
1.632.11
2026.05
1.55-
2026.05
1.52-
2026.05
1.5-
2026.05
1.43-
2026.05
1.43-
2026.05
1.43-
2026.05
1.36-
2026.05
1.36-
2026.05
1.35-
2026.05
1.31-
2026.05
1.31-
2026.05
1.29-
2026.05
1.28-
2026.05
1.22-
2026.05
11
2026.05
11
2026.05
11
2026.05
11
2026.05
11
2026.05
11
2026.05
-3.84
2026.05
-3.9