Deep Research
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
78.4Score
68
Jun 9, 2026
58.03RACE Overall
58
Jul 2, 2026
58.1Pass@1
45
Feb 26, 2026
79.2Score
42
Jun 15, 2026
59.5Score
38
Jun 15, 2026
55.33Accuracy
38
Mar 25, 2026
50.9Pass@1
33
Mar 27, 2026
83Accuracy
30
Feb 26, 2026
0.508RACE Overall
24
Feb 26, 2026
78.2Accuracy
24
Mar 24, 2026
82Score
22
Apr 23, 2026
62.8Average Score
21
Jun 15, 2026
50.7Overall Score
21
Jun 15, 2026
49.7Pass@1
20
Feb 26, 2026
74.1Pass@1
20
Feb 26, 2026
88.3Score
18
May 18, 2026
66.6Accuracy
18
Feb 26, 2026
80.1Score
17
Apr 23, 2026
50.2Average Score
17
Apr 10, 2026
81.3BrowseComp-zh Score
16
Feb 26, 2026
51.46Pass@1
16
Feb 26, 2026
51Accuracy
16
Feb 26, 2026
71Pass@1
15
Feb 26, 2026
58.1Pass@1
15
Feb 26, 2026
70.5Pass@1
15
Feb 26, 2026