Deep Research on xBench-DS-2505
82ScoreSMTL-30B-300
Evaluation Results
| Method | Links | |
|---|---|---|
| SMTL-30B-300Model Category=Trained Agents, Model Scale Group=≥30B2026.04 | 82 | |
| DeepSeek-V3.2Model Category=Foundation Models2026.04 | 78 | |
| GPT-5 HighModel Category=Foundation Models2026.04 | 77.8 | |
| Tongyi-DR-30BModel Category=Trained Agents, Model Scale Group=≥30B2026.04 | 75 | |
| DR-Venus-4B-RLModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=RL2026.04 | 74.7 | |
| OpenSeeker-v1-30B-SFTModel Category=Trained Agents, Model Scale Group=≥30B, Training Protocol=SFT2026.04 | 74 | |
| WebSailor-V2-30B-RLModel Category=Trained Agents, Model Scale Group=≥30B, Training Protocol=RL2026.04 | 73.7 | |
| GLM-4.7Model Category=Foundation Models2026.04 | 72 | |
| AgentCPM-Explore-4BModel Category=Trained Agents, Model Scale Group=≤9B2026.04 | 70 | |
| DR-Venus-4B-SFTModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=SFT2026.04 | 69 | |
| MiniMax-M2.1Model Category=Foundation Models2026.04 | 68.7 | |
| OpenAI-o3Model Category=Foundation Models2026.04 | 67 | |
| OpenResearcher-30B-A3BModel Category=Trained Agents, Model Scale Group=≥30B2026.04 | 65 | |
| DeepMiner-32B-RLModel Category=Trained Agents, Model Scale Group=≥30B, Training Protocol=RL2026.04 | 62 | |
| WebExplorer-8B-RLModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=RL2026.04 | 53.7 | |
| DeepDive-32B-RLModel Category=Trained Agents, Model Scale Group=≥30B, Training Protocol=RL2026.04 | 50.5 | |
| OffSeeker-8B-DPOModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=DPO2026.04 | 49 | |
| DeepDive-32B-SFTModel Category=Trained Agents, Model Scale Group=≥30B, Training Protocol=SFT2026.04 | 48.5 | |
| OffSeeker-8B-SFTModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=SFT2026.04 | 48 | |
| DeepDive-9B-RLModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=RL2026.04 | 38 | |
| DeepDive-9B-SFTModel Category=Trained Agents, Model Scale Group=≤9B, Training Protocol=SFT2026.04 | 35 | |
| WebSailor-7BModel Category=Trained Agents, Model Scale Group=≤9B2026.04 | 34.3 |