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

Deep Search on BrowseComp (test)

49.7Accuracy

OpenAI-o3

3.4215.43527.4539.465Jan 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
49.7
2026.01
45.1
2026.01
43.4
2026.01
30
2026.01
26.4
2026.01
24.8
2026.01
20.5
2026.01
18.9
2026.01
18.5
2026.01
17.9
2026.01
17.5
2026.01
16
2026.01
15.3
2026.01
15.2
2026.01
14.7
2026.01
14.5
2026.01
14.1
2026.01
13.9
2026.01
13.1
2026.01
12.9
2026.01
12.2
2026.01
11.5
2026.01
10.5
2026.01
7.7
2026.01
6.7
2026.01
6.3
2026.01
5.2