<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Docling on René Welches</title><link>https://blog.renewelches.com/tags/docling/</link><description>Recent content in Docling on René Welches</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Tue, 26 May 2026 15:09:02 -0400</lastBuildDate><atom:link href="https://blog.renewelches.com/tags/docling/index.xml" rel="self" type="application/rss+xml"/><item><title>My Local AI Stack: Mac Mini + Proxmox LXCs</title><link>https://blog.renewelches.com/2026/05/26/local-ai-stack-mac-mini-proxmox/</link><pubDate>Tue, 26 May 2026 10:00:00 +0000</pubDate><guid>https://blog.renewelches.com/2026/05/26/local-ai-stack-mac-mini-proxmox/</guid><description>&lt;p&gt;I have been slowly assembling a local AI setup that I actually want to use day to day. The goal was simple: a Claude/ChatGPT-like experience that runs on my own hardware, with web search and document RAG, and without sending my prompts off to someone else&amp;rsquo;s servers. It ended up being a two-machine split — a Mac mini doing the inference, and a couple of Proxmox LXCs doing the frontend work — and I am surprisingly happy with how it landed.&lt;/p&gt;</description></item></channel></rss>