Tool Calling
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
94.84Score (Name)
16
Apr 8, 2026
63Accuracy
13
May 19, 2026
73.8Accuracy
12
Jun 11, 2026
92.5Accuracy
12
Mar 4, 2026
94.84F1 Name
12
Feb 26, 2026
98.14F1 Name
12
Feb 26, 2026
87.63F1 Name
12
Feb 26, 2026
88F1 Name Match
12
Feb 26, 2026
90.78F1 Score
12
Apr 30, 2026
45.6Task Goal Completion (TGC)
11
Jun 12, 2026
63.1Multi-turn Success Rate
11
Jun 2, 2026
70.4Accuracy
10
May 14, 2026
26Exact Match (EM)
10
Apr 14, 2026
57.24Exact Match (EM)
10
Apr 14, 2026
74.05EM
10
Apr 14, 2026
0.02smECE
9
Jul 2, 2026
78.72TSA
9
Apr 23, 2026
82.28TSA
9
Apr 23, 2026
88.47F1 (Name)
9
Feb 26, 2026
9.1Task Goal Completion (TGC)
8
Jun 11, 2026
79Full Accuracy
8
May 27, 2026
62.5Accuracy
8
May 19, 2026
34.7Task Completion Rate
8
Apr 21, 2026
25Multistep (Agent)
7
Jun 2, 2026
-1Δ Accuracy
7
May 12, 2026