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

WideSearch

Benchmarks

Task NameDataset NameSOTA ResultTrend
Broad Information SeekingWideSearch
Item F1 (Avg@4)80.12
34
Information RetrievalWideSearch 2025
Item F1 Avg@467.81
20
Structured Information SynthesisWideSearch English
Success Rate (Avg @ Depth 4)8.38
18
Wide SearchWideSearch 40 samples
ReAct Acc9.5
14
Long-horizon agent performanceWideSearch
Overall Score72.7
13
Web ResearchWideSearch
Accuracy74.2
11
Web Search ResearchWideSearch
Score76.2
7
Agentic SearchWideSearch
Pass@165.7
5
Broad information seekingWideSearch English
Item F162
5
Showing 9 of 9 rows