Setting up Rocky Linux 10 on a GMKtec EVO-X-2 with llama.cpp
Setting up a local AI inference box using a GMKtec EVO-X-2 mini PC running Rocky Linux 10 and llama.cpp.
Sr. Customer Support Engineer @ CIQ | 10+ Years in Linux & Open Source
Live Last read · last published · next check

Running Google's Gemma 4 31B locally on the GMKtec EVO-X-2 to assist with woodworking projects — from timber selection and joinery planning to finish schedules and cut lists.

Maximizing the utility of a 96GB GMKtec EVO-X-2 by running multiple concurrent llama.cpp instances of Gemma 4 31B with full 128K context.
Setting up a local AI inference box using a GMKtec EVO-X-2 mini PC running Rocky Linux 10 and llama.cpp.
I have been using Claude daily for Linux troubleshooting and open-source work for a while now, so I wanted to document my current setup. Here is what I am running.

Miyashita Park is a skate park located in the heart of Shibuya, Tokyo ( Google Maps ). I recently visited and had a fantastic time there, so I wanted to share my experience. Pricing and Entry If you live within Shibuya, the cost for adults is 500 yen. For those living outside of Shibuya or for skateboarding tourists, the cost is 1,000 yen for 2 hours. You purchase a wristband at the park office…
I have been studying n8n over the past couple of weeks and I am genuinely impressed by how powerful it is. For those unfamiliar, n8n is an open-source workflow automation platform that lets you connect different services, APIs, and AI models together into visual workflows. Think of it as a canvas where you drag and drop nodes, wire them together, and suddenly you have a fully automated pipeline…
Welcome to the AI Journey section of my site. This is where I’ll be documenting my activities and projects involving artificial intelligence.