OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data
Open paper
Benchmark
Model
Rank
Results
deep-research-agents-on-browsecomp-zh
OpenSeeker-v1-30B-SFT
#8
Accuracy: 48.4
deep-research-agents-on-browsecomp
OpenSeeker-v1-30B-SFT
#56
Accuracy: 29.5
Rank counts only results with a code link.