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

DeepSearchQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
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
Agentic SearchDeepSearchQA
Accuracy90
14
Deep ResearchDeepSearchQA
Score80
9
Search-based Question AnsweringDeepSearchQA
Pass64
4
Showing 8 of 8 rows