The "model-provider" topic on GitHub groups 21 open-source projects in the DeepSeek Harness (DSH) ecosystem, led by dsh-our-free-model with 19.3k GitHub stars. dsh-our-free-model — 在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 DeepSeek V4. Every project here is indexed by DSH Universe with live GitHub data — stars, activity and install status — so you can compare and install directly.
在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 DeepSeek V4.1 Flash、Kimi K3 在内的前沿模型——完全免费,不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, DeepSeek V4.1 Flash and Kimi K3 among them. Completely free, with no usage cap.
Use your ChatGPT subscription as a model provider in DeepSeek Harness (dsh). Sign in with ChatGPT over OAuth — no API key — and the account's own models appear in the model picker.
Add Low, Medium, High and other reasoning levels to DeepSeek Harness third-party APIs, with model selection for each subagent call | Add Low, Medium, High, and other reasoning levels to third-party APIs, with model selection for each subagent call
DeepSeek Harness third-party API and custom model settings plugin: supports request headers, User-Agent, model list, image input and reasoning levels | WebUI plugin for third-party APIs and custom models with request headers, image input, and reasoning levels
DeepSeek Harness (DSH) plugin: Mistral AI subscription (Vibe Pro/Team) model provider with multi-account failover, prompt-cache tuning and a composer usage panel. · DeepSeek Harness(DSH)插件:接入 Mistral AI 订阅(Vibe Pro/Team)的模型提供方,支持多账号故障切换、提示词缓存调优和输入框额度面板。
DeepSeek Harness plugin — pre-probed OpenCode Go / Zen model routes with availability annotations (region blocks, no credit, client-only free tiers) written into the model name.
CatPaw (妙手) model bridge for DeepSeek Harness: 12 real models, per-model reasoning effort, harness-executed tool calling, and QR login without the desktop app.
为 DSH 接入 AMD Radeon Cloud 模型 provider,支持逐模型设置思考档位 | Adds AMD Radeon Cloud as a model provider for DeepSeek Harness with per-model reasoning effort levels.