List your company

Open-ended Deep Research on Composite Suite (ScholarQA, HealthBench, ResearchQA, DeepResearchBench v2)

80.2ResearchQA AccuracySearchSwarm
55.13661.64368.1574.657Jun 8, 2026

Evaluation results

MethodLinks
SearchSwarmSystem Category=Open-source models2026.0680.279.252.844.464.2
OpenAI DeepResearchSystem Category=Closed-source systems2026.0679.279.653.846.964.9
Dr.TuluSystem Category=Open-source models2026.0675.788.352.845.465.6
Perplexity DeepResearchSystem Category=Closed-source systems2026.0675.367.3—42.3—
Gemini-3.1-Pro + searchSystem Category=Closed-source systems2026.0674.5—47.544.4—
WebThinker-32B-DPOSystem Category=Open-source models2026.0674.246.739.440.650.2
Tongyi DeepResearchSystem Category=Open-source models2026.0666.746.546.240.650
QwQ-32BSystem Category=Open-source models2026.0660.941.924.540.341.9
Qwen3-8BSystem Category=Open-source models2026.0656.140.416.533.336.6