Deep-search QA on BrowseComp (test)
51.5Pass@1OpenAI-DeepResearch
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenAI-DeepResearchType=Proprietary Agentic Foundation Model2026.02 | 51.5 | |
| OpenAI-o3Type=Proprietary Agentic Foundation Model2026.02 | 49.7 | |
| GLM-4.6Type=Open-Source Agentic Foundation Model2026.02 | 45.1 | |
| Tongyi-DR-30BBackbone=30B, Type=Open-Source Agentic Foundation Model2026.02 | 43.4 | |
| MiroThinker-v1.0-30BBackbone=30B, Type=Open-Source Agentic Foundation Model2026.02 | 41.2 | |
| DeepSeek-V3.1Type=Open-Source Agentic Foundation Model2026.02 | 30 | |
| Claude-4.5-SonnetType=Proprietary Agentic Foundation Model2026.02 | 19.6 | |
| QwQ-32B-RFT-Iter2 (SRR-Judge)Backbone=QwQ-32B, Training Strategy=Iterative RFT, Iteration=2, Supervision Type=SRR-Judge2026.02 | 16.2 | |
| DeepDive-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 15.3 | |
| QwQ-32B-RFT-Iter1 (SRR-Judge)Backbone=QwQ-32B, Training Strategy=Iterative RFT, Iteration=1, Supervision Type=SRR-Judge2026.02 | 15.1 | |
| DeepSeek-R1 + SRR RefineBackbone=DeepSeek-R1, Refinement Method=SRR Refine2026.02 | 14.6 | |
| QwQ-32B-SFT + SRR RefineBackbone=QwQ-32B, Training Strategy=SFT, Refinement Method=SRR Refine2026.02 | 14.4 | |
| QwQ-32B-RFT-Iter1 (ORM)Backbone=QwQ-32B, Training Strategy=Iterative RFT, Iteration=1, Supervision Type=Outcome Supervision (ORM)2026.02 | 13.9 | |
| QwQ-32B-SFTBackbone=QwQ-32B, Training Strategy=SFT2026.02 | 12.6 | |
| AFM-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 11.1 | |
| WebSailor-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 10.5 | |
| QwQ-32B + SRR RefineBackbone=QwQ-32B, Refinement Method=SRR Refine2026.02 | 9.1 | |
| Kimi-K2-0905Type=Open-Source Agentic Foundation Model2026.02 | 7.4 | |
| DeepSeek-R1Backbone=DeepSeek-R12026.02 | 7.3 | |
| QwQ-32BBackbone=QwQ-32B2026.02 | 5.4 | |
| ASearcher-Web-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 5.2 | |
| WebDancer-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 3.8 | |
| Search-O1-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 2.8 | |
| WebThinker-32BBackbone=32B, Type=Open-Source Deep Search Solution2026.02 | 2.8 |