Top tier post. Can’t wait for the experts at awful.systems to come explain. The question by OP is a very good question so I am sure we will get expert explanations from users of the awful.systems instance.

If you would rather not trust any AI company with your data, consider heading to !localllama@sh.itjust.works where self-hosted LLM are discussed!

I see. Mistral was the favorite in self-hosted LLM circles back in 2023-2024 but general opinion is that they have since been far surpassed by Chinese and American models, hence my question.

Good to know they’ve found a market with their online offering.

Is there a use case where Mistral still beat Qwen or Gemma? If you’re using Mistral, which model and what do you use it for?

It is hardcoded. The string ‘enoughmuskspam’ is right here https://codeberg.org/rimu/pyfedi/src/commit/b7a9ea0eea3a80f710e0b5b63cf0bbecde60f8bf/app/admin/routes.py#L373

I have noted in other threads and will note again here because people keep attacking me about this: hardcoding does not mean the behavior is not circumventable. It just means the string is in the source file (rather than a config file or database).