Chat
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
8.91MT-Bench Score
73
Jul 9, 2026
57.46AlpacaEval (LC win %)
58
May 19, 2026
3,213Win Rate
39
Mar 4, 2026
8MT-Bench Score
19
Feb 26, 2026
1.79Success Rate (SR)
16
May 20, 2026
84.3LC Win Rate
16
May 18, 2026
48.8Loose Prompt Metric
15
Feb 26, 2026
3.59Speedup
12
May 29, 2026
2,900English Score
9
Feb 26, 2026
2.85Speedup
8
Jul 9, 2026
27.86AlpLC Score
8
May 12, 2026
49.3Chat Score
8
Feb 26, 2026
7.89Overall Score
2
Feb 26, 2026
8.69Score
2
Feb 26, 2026