If you've got a regular consumer GPU and have been curious about running a local LLM, there's a good chance you've either ...