I have a question about llama.cpp that may be useful for other people working with it too.
I am particularly interested in privacy considerations and reliability. I would like to compare approaches that people have actually used rather than just theoretical suggestions.
Are there any settings...
