Selfhosted

Currently I've been using a local AI (a couple different kinds) to first - take the audio from a Twitch stream; so that I have context about the conversation, convert it to text, and then use a second AI; an LLM fed the first AIs translation + twitch chat and store 'facts' about specific users so that they can be referenced quickly for a streamer who has ADHD in order to be more personable.

That way, the guy can ask User X how their mothers surgery went. Or he can remember that User K has a birthday coming up. Or remember that User G's son just got a PS5 for Christmas, and wants a specific game.

It allows him to be more personable because he has issues remembering details about his users. It's still kind of a big alpha test at the moment, because we don't know the best way to display the 'data', but it functions as an aid.

I think that's a size where it's a bit more than a good autocomplete. Could be part of a chain for retrieval augmented generation. Maybe some specific tasks. And there are small machine learning models that can do translation or sentiment analysis, though I don't think those are your regular LLM chatbots... And well, you can ask basic questions and write dialogue. Something like "What is an Alpaca?" will work. But they don't have much knowledge under 8B parameters and they regularly struggle to apply their knowledge to a given task at smaller sizes. At least that's my experience. They've become way better at smaller sizes during the last year or so. But they're very limited.

I'm not sure what you intend to do. If you have some specific thing you'd like an LLM to do, you need to pick the correct one. If you don't have any use-case... just run an arbitrary one and tinker around?

absolutely nothing

Nothing.