Llamafile is great but solves a slightly different problem very well: how do I easily download and run a single model without having any infrastructure in place first?

Ollama solves the problem of how I run many models without having to deal with many instances of infrastructure.

You don't need any infrastructure for llamafiles, you just download and run them (everywhere).

Yes, that's what I meant, sorry if it was confusing: The problem that Llamafiles solve is making it easy to set up one model without infrastructure.

It's actually more difficult to use on linux (compared to ollama) because of the weird binfmt contortions you have to go through.

What contortions? None of my machines needed more than `chmod +x` for llamafile to run.