The open-weight space moves fast — those who’ve spent any time running local LLMs can attest. The ambitious project of running a language model on your own hardware is now a weekend afternoon project.