BrowseComp
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BrowseComp ZH
82.4Score
17
BrowseComp-VL
45.2L1 Score
17
BrowseComp
71.2Accuracy
17
BrowseComp-ZH
64.2Overall Score
17
BrowseComp (BC)
51.5Pass@1 Accuracy
17
BrowseComp-Plus
13Recall@5
17
Browsecomp VL
54.9Accuracy
17
BrowseComp+ Complex Tasks 1st Pass
90Accuracy
16
BrowseComp
67.6Accuracy
16
BrowseComp complete (test)
88.2Avg Success Rate@3
16
BrowseComp-zh
81.3BrowseComp-zh Score
16
BrowseComp-Plus
71.2Accuracy
15
BrowseComp
59.3Accuracy
15
Browsecomp
51.5Accuracy
15
BrowseComp EN
67.6Score
15
BrowseComp Our Settings
62.5Pass@1
15
BrowseComp (BC)
67.8Pass@3
15
BrowseComp-ZH
58.1Pass@1
15
BrowseComp-ZH
82.4Score
15
BrowseComp-Plus
66.5Accuracy
14
BrowseComp
21String-F1
14
BrowseComp-ZH
76.1Score
13
BrowseComp-Plus 100k corpus (100-query sample)
82Accuracy
13
BrowseComp Full
56.1Pass@1
13
BrowseComp-Plus
70Score
13