DeepSeek Harness QoL plugin for the local Qwen3.8 line (27B/Flash-Next): per-request thinking budgets, a compaction backend that stops burning the output cap on thinking, and a settings tab. 本地 Qwen3.8 线的 DSH QoL 插件:逐请求 thinking 预算、不再把输出帽烧在 thinking 上的压缩后端、设置 tab。
为DSH接入本地大模型能力:在「设置→插件」页一键启停本地 llama.cpp 大模型(双槽x双模态x双预设),卡片内配置、一条命令安装、自动注册,装完即用| Enable local large model capabilities for DSH: One-click start/stop for local llama.cpp models (dual‑slot × dual‑modal × dual‑preset) right in Settings → Plugins; configure within the card, install with a single command, automatically register, and ready to use
DSH (DeepSeek Harness) plugin: delegate read-only collection tasks to a local Ollama model — zero cloud token cost, zero API key. 把只读采集任务交给本机 Ollama 模型:含设置面板 + 选型尺 + AI 可读说明书。