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

Dialogue Evaluation on MT-Bench (τ and Speedup)

4.01Kendall's Tau (τ)

EAGLE-3

2.32522.76263.23.6374Feb 27, 2026Mar 20, 2026Apr 11, 2026May 2, 2026May 24, 2026Jun 14, 2026Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.02
4.013.01
2026.02
42.98
2026.02
42.22
2026.02
3.962.22
2026.02
3.953
2026.02
3.892.89
2026.02
3.882.14
2026.02
3.842.62
2026.02
3.82.64
2026.02
3.792.61
2026.02
3.782.63
2026.02
3.772.61
2026.02
3.762.82
2026.02
3.762.85
2026.02
3.752.6
2026.07
3.72-
2026.07
3.7-
2026.02
3.672.53
2026.02
3.661.85
2026.07
3.64-
2026.02
3.542.45
2026.02
3.532.37
2026.02
3.512.3
2026.02
3.52.3
2026.07
3.49-
2026.02
3.482.39
2026.02
3.442.42
2026.02
3.392.29
2026.02
3.362.13
2026.02
3.352.36
2026.02
3.351.42
2026.02
3.342.3
2026.02
3.332.09
2026.02
3.311.41
2026.02
3.292.1
2026.02
3.281.35
2026.02
3.271.41
2026.02
3.21.35
2026.07
3.19-
2026.02
3.181.98
2026.02
3.132.5
2026.02
3.121.3
2026.02
3.112.35
2026.02
3.111.96
2026.07
3.11-
2026.07
3.1-
2026.02
3.082.09
2026.07
3.07-
2026.02
3.062.68
2026.02
3.061.94
2026.07
2.98-
2026.02
2.972.22
2026.02
2.961.84
2026.02
2.921.85
2026.02
2.91.26
2026.02
2.91.66
2026.02
2.882.1
2026.02
2.861.62
2026.02
2.821.54
2026.02
2.811.96
2026.02
2.811.57
2026.02
2.81.59
2026.02
2.772.07
2026.02
2.691.48
2026.02
2.671.83
2026.02
2.671.83
2026.02
2.671.49
2026.07
2.66-
2026.02
2.641.63
2026.02
2.631.12
2026.07
2.62-
2026.02
2.561.61
2026.02
2.531.2
2026.07
2.39-