Multi-turn Dialogue Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
62.7Overall Score
532
May 26, 2026
6.34Score
90
Feb 26, 2026
98.16Performance (%)
16
May 26, 2026
8.64R1 Score
14
Jun 23, 2026
86.3Success Rate (N=16)
9
Feb 26, 2026
7.34Expansion Score
9
Feb 26, 2026
6.26MT-Bench Score
6
Mar 4, 2026
8.99MT-Bench Score
5
Feb 26, 2026
1.9Writing Score
3
Feb 26, 2026
1.79Speedup
2
Feb 26, 2026