BrowseComp
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BrowseComp
85.9Accuracy
68
BrowseComp
78.4Score
68
BrowseComp-ZH
75.9Pass@1
52
BrowseComp
67.6Pass@1
47
BrowseComp-ZH (BC-zh) original (test)
58.1Pass@1
45
BrowseComp zh
52.9Accuracy (%)
39
BrowseComp+
55.33Accuracy
38
BrowseComp-Plus (BCP) (test)
10.6Accuracy
35
BrowseComp-ZH
66.6Accuracy
35
BrowseComp-VL
49.1Accuracy (BrowseComp-VL)
34
BrowseComp-zh
83.4Accuracy
34
BrowseComp
50.9Pass@1
33
BrowseComp-Plus
83.92Accuracy
31
BrowseComp
67.6Accuracy
29
BrowseComp-Plus (test)
27.38Accuracy (%)
28
BrowseComp (test)
49.7Accuracy
27
BrowseComp
78.4Score
27
BrowseComp (test)
48.7Pass@1
25
BrowseComp-Plus
79.33Accuracy
25
BrowseComp-Plus
88.33Accuracy (Avg)
25
BrowseComp-VL
54.9Pass@1
24
BrowseComp-Plus
77.33Performance
24
BrowseComp
71.33Performance
24
BrowseComp (test)
51.5Pass@1
24
Browsecomp
52Accuracy
24