Agentic Tool-use
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
90.4Retail Score
59
Jun 2, 2026
63.5Pass@1
30
May 25, 2026
82Pass@1
30
May 25, 2026
70.3Pass@1
22
May 7, 2026
3.95Overall Score
20
Jun 12, 2026
3.54Overall Score
20
Jun 12, 2026
83.7TGC
20
Apr 6, 2026
89.3Task Goal Completion (TGC)
20
Apr 6, 2026
85.8Average Success Rate
20
Jul 9, 2026
100Accuracy
18
May 19, 2026
85.79Overall Success Rate
17
Feb 26, 2026
37.5ACE-E Score
16
Jun 1, 2026
94.1BFCL Average Score
16
May 13, 2026
53.7TS-M Score
16
May 13, 2026
71.3Retail Score
13
Feb 26, 2026
90.4Accuracy
10
Jul 2, 2026
0.775Retail Score
9
Feb 26, 2026
72.8Avg@8
8
Mar 31, 2026
73.68Avg@8
8
Mar 31, 2026
58Avg@8
8
Mar 31, 2026
85.9Pass@1
6
Jun 30, 2026
53.5Pass@1
6
Jun 30, 2026
99Success Rate
5
Jun 9, 2026
67.5Accuracy
5
Apr 13, 2026
55.36Success Rate
5
Feb 26, 2026