# Self-Hosted AI

OpenWorker is Andrew Ng's MIT-licensed, local-first desktop agent that returns finished deliverables — a drafted report, a triaged inbox, a calendar fix — instead of chat. A 5,000-word dive into its architecture, the typed permission engine, the model matrix, how it differs from cloud agents, what it costs, and where your data goes.

"DeepSeek V4-Flash is MIT-licensed, runs on a consumer GPU, and hits 79% on SWE-bench with a 1M-token context. A deep, 5,000-word look at what it is, how MoE works, what it costs to self-host, what it is honest about, and where your data goes."

Open WebUI turns a local Ollama instance into a multi-user ChatGPT your whole household can use — with RAG, document upload, image generation, voice, functions, and a plugin pipeline. MIT-licensed, runs offline, your data never leaves the box. A 5,000-word look at installing it, tuning it, real workflows, and where your data actually goes.

Ollama turns 'run a 27B model on your laptop' into a single command. MIT-licensed, OpenAI-compatible on :11434, and your data never leaves the box. A 5,000-word dive into what it is, how to tune it, quantization math, choosing your first model, a real day with it, troubleshooting, cost, and where your data goes.