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

DeepSearchQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Agentic SearchDeepSearchQA
Accuracy90
29
Question AnsweringDeepSearchQA
Accuracy82.67
19
DeepSearchQADeepSearchQA
Accuracy76.67
19
Long-horizon agentic taskDeepSearchQA
Performance66
18
ReasoningDeepSearchQA 2026 (test)
Score24.1
15
ReasoningDeepSearchQA
Score24.1
15
Deep ResearchDeepSearchQA
Score80
9
Search Agent CapacityDeepSearchQA
Score77.4
8
Search-based Question AnsweringDeepSearchQA
Pass64
4
Showing 9 of 9 rows