QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks
Open paper
Benchmark
Model
Rank
Results
agents-on-gaia-text-only-103-validation-tasks
QUEST-35B
#2
Accuracy: 80.8
deep-research-agents-on-browsecomp-plus
QUEST-35B
#10
Accuracy: 69.5
deep-research-agents-on-browsecomp
QUEST-35B
#35
Accuracy: 64.6
Rank counts only results with a code link.