List your company

Deep Research on DeepResearchGym (test)

80.67Focus StabilitySTEER
66.52670.19873.8777.542May 22, 2026

Evaluation results

MethodLinks
STEERBase Pause Cost (C0)=0.12026.0580.6760.191.1
o4-mini-deep-researchPersona information level=full-persona2026.0580.666.451.94
OpenDeepResearchPersona information level=full-persona2026.0578.8457.060.81
STEERBase Pause Cost (C0)=0.72026.0578.5160.471.13
o4-mini-deep-researchPersona information level=initial-persona2026.0578.4167.361.74
OpenDeepResearchPersona information level=initial-persona2026.0575.6157.070.81
OpenDeepResearch2026.0573.1560.360.69
GPT-ResearcherPersona information level=full-persona2026.0570.7859.941.04
GPT-ResearcherPersona information level=initial-persona2026.0569.1860.821.11
GPT-Researcher2026.0567.0761.391.02