Continuing building out my AI lab to test Harnesses and running local models. I am running 3 inference engines (LM Studio, Ollama and vLLM) so I can have 3 local mid size models running at all time. Dropped Agent Zero as it only works with cloud models. Running Craft Agents and Hermes.
Been in data centers most of my life. As models get better, I have 64G Mac and nVidia Spark 128G. Lots of people on Youtube the last 2 weeks talking about local models.
man i wanna start tuning these models bruh and like damn bruh this shit fucking sucks everything is so expensive dude... any suggestions? i wanna start tuning in 27b model's full weights one day but i know i gotta start small at like 17b maybe
@Luke Brinton Best Local LLMs for Mac in 2026: The Ultimate M1 to M6 Guide https://python.plainenglish.io/best-local-llms-for-mac-in-2026-the-ultimate-m1-to-m6-guide-2ba2ee8a829e?sk=a29ac09925918178975c373777964e6d
@Luke Brinton I have used Facebook Marketplace to find used Macs locally, prices may be cheaper until the local person try to get ebay prices. Going to ebay, the laptops I listed have a 1 year warranty and a return policy.