Generative artificial intelligences require a secondary program to access them. I've grown comfortable with a few of these. In many cases, they're produced by the same corporations that programmed grew created the LLMs. Are you surprised? I wasn't.
If you visit ChatGPT on the web, for example, Sam Altman’s people control both the LLM you speak with and the interface through which you speak with it. In that and many other cases, there are other ways of doing it, though
I don’t know terribly much about all this yet, but have been using a LLM frontend called SillyTavern. I primarily use Deepseek with the former for character dialogues. I’m learning how OpenRouter works and how to change the settings on an LLM so that you get the best possible results, and how prompting actually works.
Here, you'll find things I've written about the systems allowing you to access generative artificial intelligence. I won’t include many guides of my own here because I haven’t got the knowledge base (yet).
I couldn't handle the normal SillyTavern interface because it was boring, so I turned it into what I call the Sunrise Tearoom. Now with downloads included to a degree...
A quick, slightly bewildered glance what completion APIs are, how OpenRouter handles them, and why text and chat modes kept behaving like different creatures for me... sometimes. Will update with more notes later?
I have a lot of fun talking to my LLM coding assistants and designing their elaborate backstories, but it's not efficient...