DeepSeek Search is strongest on factual questions and weakest on multi-hop browsing. It scores 70% on AA-Omniscience, 3 points behind the leaders, 75 F1 on DeepSearchQA and 63% on BrowseComp.
DeepSeek Search is strongest on factual questions and weakest on multi-hop browsing. It scores 70% on AA-Omniscience, 3 points behind the…
DeepSeek Search is strongest on factual questions and weakest on multi-hop browsing. It scores 70% on AA-Omniscience, 3 points behind the leaders, 75 F1 on DeepSearchQA and 63% on BrowseComp.
Artificial Analysis · X
Content notes
Original publication and rights belong to the source.