Deep Research
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
71Pass@1
15
Feb 26, 2026
56.28Comprehensiveness
14
May 26, 2026
72.2Pass@1
14
Feb 26, 2026
67Accuracy
14
Mar 24, 2026
56Accuracy
14
Feb 26, 2026
57.58Information Recall
13
Jun 8, 2026
32.9Pass@1
13
Feb 26, 2026
46.45Overall Score
12
Feb 26, 2026
75Score
11
Apr 23, 2026
80.67Focus Stability
10
May 26, 2026
70.5Average Score
10
Apr 6, 2026
28.3Accuracy
10
Mar 24, 2026
60.01Mean@3
9
Jun 1, 2026
80Score
9
Apr 23, 2026
2,535Score Range
9
Feb 26, 2026
50.8Score
8
Jul 2, 2026
61.1Overall Score
8
Jul 2, 2026
51.9Overall Score
8
May 13, 2026
53.4Average Score
8
Apr 29, 2026
620Mean Correct Answer Rate
8
Feb 26, 2026
49.2Mean Correct Rate
8
Feb 26, 2026
17.6Mean Correct Answer Rate
8
Feb 26, 2026
92.8Mean Answer Rate
8
Feb 26, 2026
81Mean Correct Answer Rate
8
Feb 26, 2026
87.9Mean Correct Rate
8
Feb 26, 2026