Deep Research on DeepConsult (test)
80Win RateFS-Researcher
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| FS-ResearcherCategory=Ours, Backbone=Claude-Sonnet-4.5, Sampling protocol=sampled 20 queries2026.02 | 80 | 10.42 | 9.58 | 8.33 | |
| FS-ResearcherCategory=Ours, Backbone=GPT-52026.02 | 73.28 | 18.97 | 7.76 | 7.26 | |
| EnterpriseDeepResearchCategory=Open Source, Backbone=Gemini-2.5-Pro2026.02 | 71.57 | 19.12 | 9.31 | 6.82 | |
| RhinoInsightCategory=Open Source, Backbone=Gemini-2.5-Pro2026.02 | 68.51 | 11.02 | 20.47 | 6.82 | |
| WebWeaverCategory=Open Source, Backbone=Qwen3-235B-A22B-Instruct-25072026.02 | 66.16 | 12.14 | 21.68 | 6.94 | |
| Gemini-2.5-Pro-DeepResearchCategory=Proprietary2026.02 | 61.27 | 31.13 | 7.6 | 6.7 | |
| Claude-DeepResearchCategory=Proprietary2026.02 | 25 | 38.89 | 36.11 | 4.6 | |
| OpenAI-DeepResearchCategory=Proprietary2026.02 | 0 | 100 | 0 | 5 |