gavinj.net

I wanted to know whether a sub-$3k Australian workstation GPU could run useful modern local LLMs. The short answer is yes, if you pick the right model, the right quant, and build around the 24GB VRAM limit instead of pretending it is a tiny cloud.

After a few days with Odysseus, I came away impressed by the web interface, local model support, and model comparison tools, but frustrated by weak Gmail-style integrations and an agent experience that never quite felt personal.

OpenClaw died on me. The gateway wouldn't start, cron jobs were imaginary, and my AI assistant had gone full radio silent. Two days with Hermes Agent on a 2GB VPS and I'm wondering why I waited this long.

Ubuntu 26.04 LTS is not trying to turn the desktop into a chatbot. It is making it easier to run AI tools locally, see what hardware is being used, and choose which models belong on your machine.

Google's Gemma 4 models are impressive, but the real story is the licensing. For the first time, a major AI lab has released capable open models under a true Apache 2.0 license. Here's why that changes everything.