Deep Research on BrowseComp-EN (BC-en) original (test)
49.7Pass@1OpenAI-o3
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenAI-o3Backbone Group=Large Size Foundation Models (> 100B)2026.01 | 49.7 | |
| Tongyi-DeepResearchBackbone Group=Medium Size Deep Research Agents (30B-72B)2026.01 | 43.4 | |
| DeepSeek-V3.2Backbone Group=Large Size Foundation Models (> 100B)2026.01 | 40.1 | |
| WebSailor-v2-30B-A3B (RL)Backbone Group=Medium Size Deep Research Agents (30B-72B), Training Protocol=RL2026.01 | 35.3 | |
| DeepSeek-V3.1Backbone Group=Large Size Foundation Models (> 100B)2026.01 | 30 | |
| WebSailor-v2-30B-A3B (SFT)Backbone Group=Medium Size Deep Research Agents (30B-72B), Training Protocol=SFT2026.01 | 24.4 | |
| WebExplorer-8B (RL)Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=RL2026.01 | 14.6 | |
| Kimi-K2Backbone Group=Large Size Foundation Models (> 100B)2026.01 | 14.1 | |
| MiroThinker-32B-DPO-v0.1Backbone Group=Medium Size Deep Research Agents (30B-72B), Training Protocol=DPO2026.01 | 13 | |
| OffSeeker-8B (DPO)Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=DPO2026.01 | 12.8 | |
| Claude-4-SonnetBackbone Group=Large Size Foundation Models (> 100B)2026.01 | 12.2 | |
| WebSailor-72BBackbone Group=Medium Size Deep Research Agents (30B-72B)2026.01 | 12 | |
| OffSeeker-8B (SFT)Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=SFT2026.01 | 10.6 | |
| WebSailor-32BBackbone Group=Medium Size Deep Research Agents (30B-72B)2026.01 | 10.5 | |
| MiroThinker-8B-DPO-v0.1Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=DPO2026.01 | 8.7 | |
| WebSailor-7BBackbone Group=Small Size Deep Research Agents (< 10B)2026.01 | 6.7 | |
| DeepDive-9B (RL)Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=RL2026.01 | 6.3 | |
| DeepDive-9B (SFT)Backbone Group=Small Size Deep Research Agents (< 10B), Training Protocol=SFT2026.01 | 5.6 | |
| ASearcher-Web-QwQBackbone Group=Medium Size Deep Research Agents (30B-72B)2026.01 | 5.2 | |
| WebDancer-QwQBackbone Group=Medium Size Deep Research Agents (30B-72B)2026.01 | 3.8 |