Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Information Seeking on DeepWide Search Benchmark
Loading...
55.9
Col-F1
Claude-Sonnet-4 (TaS)
38.636
43.118
47.6
52.082
Feb 6, 2026
Col-F1
Item-Precision
Updated 4d ago
Evaluation Results
Method
Method
Links
Col-F1
Item-Precision
Claude-Sonnet-4 (TaS)
ReAct Type=MA
2026.02
55.9
63.5
Claude-Sonnet-4 (TaS) + 32B Sub-Agent
ReAct Type=MA
2026.02
52.7
67.7
Gemini DeepResearch
2026.02
51.2
58.3
Claude-Sonnet-4
ReAct Type=SA
2026.02
39.5
35.2
Claude-Sonnet-4
ReAct Type=MA
2026.02
39.3
44.2
Feedback
Search any
task
Search any
task