How do you improve data quality when working with llama.cpp?

[alicewol]

There are several ways to approach llama.cpp, so I wanted to compare experiences.

I am particularly interested in data quality and prompt design. I would like to compare approaches that people have actually used rather than just theoretical suggestions.

Are there any settings, tools, or workflow changes that made a noticeable difference?