Function Calling
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
93.4ROUGE-L
12
Apr 8, 2026
84.79F1 Score
11
Apr 8, 2026
94.94F1 Score
11
Apr 8, 2026
63.34Overall Accuracy
11
May 28, 2026
90.8Non-live AST Score
11
Feb 26, 2026
53Summary Score
11
Feb 26, 2026
72.4Base Accuracy
10
Jun 11, 2026
77.25Simple Accuracy
9
Apr 10, 2026
50Multi-Turn Overall Accuracy
9
Feb 26, 2026
57.8Overall Score
9
Feb 26, 2026
43Avg@8
9
Feb 26, 2026
83.8Accuracy
8
Jul 9, 2026
69.7Accuracy
8
Jul 9, 2026
76.8Accuracy
8
Jul 9, 2026
85.5Accuracy
8
Jul 9, 2026
71.5Accuracy
8
Jul 9, 2026
76.6Accuracy
8
Jul 9, 2026
88.74Average Accuracy
8
Jun 16, 2026
77Success Rate (SR)
8
Jun 26, 2026
87.6Overall Score
8
May 29, 2026
94Best@3
8
May 22, 2026
78AUROC
8
Apr 28, 2026
81AUROC
8
Apr 28, 2026
0.76AUROC
8
Apr 28, 2026
77AUROC
8
Apr 28, 2026