Local LLMs are usually something that gets associated with desktop workflows, especially considering how resource-hungry they can be. So that's pretty much what I defaulted to for my local AI work, ...
Claude, Gemma4, a few Excel sheets, and vibe-coded duct tape ...