Deep Research on xbench-DS
71Pass@1DeepSeek-V3.1
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-V3.1Model Category=Advanced Models2025.10 | 71 | — | |
| OpenAI-o3Model Category=Advanced Models2025.10 | 66.7 | — | |
| Claude-4-SonnetModel Category=Advanced Models2025.10 | 64.6 | — | |
| WebSailorModel Size=32B2025.10 | 53.3 | — | |
| Kimi-K2Model Category=Advanced Models2025.10 | 50 | — | |
| Web-30B-E-GRPOModel Size=30B, Training Strategy=E-GRPO2025.10 | 46.7 | 66 | |
| Web-30B-GRPOModel Size=30B, Training Strategy=GRPO2025.10 | 45.3 | 65 | |
| Web-30B-SFTModel Size=30B, Training Strategy=SFT2025.10 | 43.7 | 63 | |
| Web-7B-E-GRPOModel Size=7B, Training Strategy=E-GRPO2025.10 | 42 | 59 | |
| Web-7B-GRPOModel Size=7B, Training Strategy=GRPO2025.10 | 40.7 | 56 | |
| WebDancer-QwQ2025.10 | 39 | — | |
| Web-7B-SFTModel Size=7B, Training Strategy=SFT2025.10 | 37.3 | 55 | |
| WebSailorModel Size=7B2025.10 | 34.3 | — | |
| WebThinker-RL2025.10 | 24 | — | |
| R1-SearcherModel Size=7B2025.10 | 4 | — |