dsh-tool-turbo¶
⭐ 7 · ✅ 活跃 · 插件
| 类型 | 插件 | 分类 | 视觉与多模态 |
| 星数 | ⭐ 7 | 状态 | ✅ 活跃 |
| 作者 | Electricitysheep | 更新时间 | — |
| 子分类 | 👁️ 视觉工具 | 能力 | coding |
一句话介绍¶
Per-round reasoning_effort optimizer for DeepSeek Harness (dsh): auto-downgrades tool-call reasoning for simple tool chains, lifting back for heavy work. Cuts thinking time between tool calls.
详细介绍¶
Cut tool-call latency in DeepSeek Harness (dsh) by auto-adjusting reasoning_effort per tool round. 中文文档 · English In a multi-step tool chain, the model re-thinks before every tool call — and that thinking dominates the wall-clock time (a 50-step agent task can spend minutes in reasoning between tools). dsh-tool-turbo watches the recent tool calls of a step and injects the lowest sensible reasoning effort into the next model request, then lifts it again the moment the work gets heavy.
🚀 快速开始¶
[tool-turbo] agent/request: baseline=high calls=[] => reasoningEffort=high
[tool-turbo] agent/request: baseline=high calls=[{"name":"write",…}] => reasoningEffort=low