Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design
Open paper
Benchmark
Model
Rank
Results
deep-research-agents-on-browsecomp-zh
Marco-DeepResearch-8B
#9
Accuracy: 47.1
deep-research-agents-on-browsecomp
Marco-DeepResearch-8B
#52
Accuracy: 31.4
Rank counts only results with a code link.