Raw LLM Responses

Locally running, less restricted models are likely to be the standard once these things start rolling out for general use. The cloud based models make for a good general purpose demonstration, but the hardware requirements to *run* the models are not as dramatic as their creators are implying. There are already folks nipping at GPT-4's heels with local models running on mid range gaming desktops. Once optimization starts to ramp up, there's no reason we couldn't all be running our own local GAI modeled to our specifications in terms of restriction and censorship. That's going to be damned hard for China to deal with.