Feature/Bug Requests

The current LLMs tend to overuse certain phrases that give away that the text has been AI-generated (so-called "AI slop"). Some backends support the banned_strings parameter which might help with that, and it would be nice to have this on Parasail too.

Download percentage in dedicated when spooling up a model would be greatly appreciated rather than guesstimating its progress.

it would be nice on the eyes if the UI had a built-in dark mode toggle :)

I have a few use cases and I would love to know what model you all think is the best based on what you've seen from your customers.

Rate limits are understandable. I just hit one, using a mix of 405b and 70b. I'd like more information on why this happened, what the rate limits are, and when they reset so I can continue on my project.