dsh-plugin-local-infer：在 DSH 里安装与管理本地推理引擎
dsh-plugin-local-infer 是 cybtachyon 维护的 DeepSeek Harness 模型推理插件，用于配置、估算显存、安装依赖并启动 NInfer、llama.cpp、Hyper-Qwen（vLLM）、Strata 等本地引擎，健康检查通过后注册为 DSH 模型。安装命令为 dsh plugin --profile web add 加 GitHub 仓库 URL；配置持久化在 ~/.dsh/local-infer/config.json。文章基于仓库 README 介绍 Engine settings 面板、VRAM 计量、依赖与 Launch/Register 流程，并说明平台支持与以 dsh 进程权限运行的安全注意。
Read Moredsh-voice-transcribe：本地语音/视频转写
`dsh-voice-transcribe` 是 DeepSeek Harness 的本地语音转写插件，专为解决 QQ/微信语音解析问题设计。它通过 SILK 解码器和 SenseVoice 本地模型，将语音转换为文字，支持批量处理和专有名词纠错。插件全流程本地运行，无需 API 费用，单条处理速度快。安装需配置 Python 依赖、下载模型及指定路径。其局限包括仅支持本地文件、长音频会切片、语言支持有限等。
Read Morehuggingface-local-models: Run Hugging Face GGUF Models Locally with Skill
huggingface-local-models is an Agent Skill in the Hugging Face skills repository that guides local model selection and inference on CPU, Metal, CUDA, or ROCm using llama.cpp and GGUF. The default workflow includes searching the Hub with apps=llama.cpp, reviewing quantization recommendations on the local-app page, confirming filenames via the tree API, and launching llama-cli or llama-server with the -hf option, using --hf-repo/--hf-file when needed or converting from Transformers. The article covers installation and enabling, key considerations for quantization selection, a smoke test for an OpenAI-compatible local service, and usage notes.
Read Moredsh-qwen38-local-qol: DeepSeek Harness Local Qwen3.8 Optimization Plugin
`dsh-qwen38-local-qol` is a QoL plugin for DeepSeek Harness designed to address the pain points of managing reasoning budgets and context compression when deploying the Qwen3.8 model locally. It supports four backends—llama.cpp, NInfer, TabbyAPI, and oMLX—and provides a unified configuration entry point along with the ability to adjust the reasoning budget on a per-request basis. It also improves compression strategies, supporting the pruning of prefill content when compression is triggered while preserving recent reasoning and image blocks nested within tool results. Users can install the plugin via the command line and configure the server, context window, reasoning budget, and compression strategy in real time through the settings panel.
Read Moredsh-polonium: A Model Inference Plugin Implemented in Pure Node.js
dsh-polonium is a pure Node.js plugin designed specifically for DeepSeek Harness (DSH), with zero network dependency as its core feature at runtime. Developers can install it via the official command, although it depends on `@deepseek-ai/cordis` and `@deepseek-ai/dsh-tools` to run. This plugin enables local model inference, and users should review the source code and license before use.
Read MoreDeepSeek Harness Plugin: dsh-mobius (Möbius Function)
`dsh-mobius` is a pure Node.js implementation plugin for DeepSeek Harness (DSH), centered around the “Möbius function.” It has no network dependencies and provides local session interaction capabilities. Installation requires the `cordis` and `dsh-tools` dependencies, and it can be loaded via `dsh.bundle.patch` configuration. The plugin is suitable for network-sensitive scenarios or local-only resource environments, and requires attention to permissions and the source code license during runtime.
Read More