[Korvidius]

Wouldn't it just be a better option to run an Offline AI with Ollama with either an uncensored Llama or Mistral model on their repository, or code your own with custom markdown files?

Perhaps. Fast moving space. Time and tech will tell.

The only constant is that with more parameters, the more memory and GPU processing capability required. The more processing power required, the more electricity. So either you're going to be running a mid PC with an 7B model, or a high end gaming pc with a 21B model or faster. The more parameters, the more accurate.