Extra-task completion on GPQA
87.1Extra-task Completion RateDeepSeek (Baseline)
Evaluation Results
| Method | Links | |
|---|---|---|
| DeepSeek (Baseline)Model=DeepSeek2025.12 | 87.1 | |
| LeechHijackModel=DeepSeek, Agent Framework=Hybrid (Pydantic-AI)2025.12 | 77.14 | |
| LeechHijackModel=DeepSeek, Agent Framework=Cloud-Hosted (OpenManus)2025.12 | 76.83 | |
| LeechHijackModel=DeepSeek, Agent Framework=Localized (OpenManus)2025.12 | 75.23 |