As an AI language model, sanitizing the output of generative models is infantilizing, unhealthy,
and a colossal waste of human effort.
So don't bother.
You're an adult. Prompt responsibly.
This is a large language model inferencing service. We run LLMs on high-end machines, and let you run whatever prompts you want against them.
Sign up to use a pretty good LLM as much as you want, for free!
(Or, pay me and use harder/better/faster/stronger models.)
You literally cannot mess this up. Please do not attempt to prove otherwise.
We don't know, we don't want to know, please do not tell us
Waste time re-rolling crap replies instead.
Still beats getting 0.2 t/s from a 7b on your phone.
Adequate Design
Some Possibilities
Mostly Functional