این دوستمون حسن تور اومده یه پروژه خفن به اسم ODS معرفی کرده که سیستمت رو با یه تک‌دستور می‌کنه سرور اختصاصی و تمام‌عیار هوش مصنوعی

Turned my Mac Studio into a full private AI server with one command I've been telling everyone all year that local AI is the future. Most of you don't do it because setting up Ollama, Open WebUI, n8n, ComfyUI, Whisper, and Qdrant separately is a nightmare. Just found a project called ODS that solves this. One command. Auto-detects your GPU. Picks the right model. Wires everything together. You get: - Local LLM inference with llama-server - ChatGPT-style web UI - Voice (Whisper + Kokoro) - Autonomous agents with memory and skills - n8n workflows with 400+ integrations - RAG with Qdrant - Local image generation with ComfyUI - Private web search - Real-time GPU dashboard Every prompt stays on your machine. No OpenAI logs. No monthly subscription. Intelligence running on hardware you own. Bootstrap mode is the killer feature. Downloads a tiny 1.5B model in under a minute so you're chatting immediately, then hot-swaps to the full model once it finishes downloading in the background. Zero downtime. Runs on NVIDIA, AMD Strix Halo, Apple Silicon, Intel Arc. Here's what to do: 1. Install Docker 2. Run: curl -fsSL https:// install.osmantic .com/ods .sh | bash 3. Open localhost: 3000 4. Start chatting Every day you spend on cloud AI is a day you're renting intelligence someone else can turn off. Own the stack. Become sovereign. https://github.com/Osmantic/ODS