Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Deep Research on BrowseComp-ZH (BC-zh) original (test)

58.1Pass@1

OpenAI-o3

5.26818.98432.746.416Jan 26, 2026Jan 28, 2026Jan 30, 2026Feb 1, 2026Feb 3, 2026Feb 5, 2026Feb 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
58.1
2026.02
58.1
2026.02
49.5
2026.01
49.2
2026.02
49.2
2026.01
47.9
2026.02
47.8
2026.01
46.7
2026.02
46.7
2026.01
44.1
2026.02
42.9
2026.02
40.8
2026.02
38.3
2026.02
37.8
2026.02
36.5
2026.02
36.5
2026.02
32.8
2026.02
32
2026.01
31.8
2026.02
31.8
2026.01
30.1
2026.02
29.7
2026.02
29.2
2026.01
29.1
2026.01
28.8
2026.01
28.3
2026.01
26.6
2026.02
26
2026.01
25.5
2026.02
25.5
2026.01
24.2
2026.02
23.4
2026.02
22.2
2026.01
18
2026.02
18
2026.02
17.9
2026.01
17
2026.01
15.7
2026.01
15.6
2026.02
15.6
2026.01
15.1
2026.01
14.2
2026.01
13.6
2026.02
12.9
2026.02
7.3