Where actually is llamafile?

[airtonix]

How to install it so i can make my own executable models?

All i see is instructions on how to use executable models.

---

ok so its over in the releases:

https://github.com/Mozilla-Ocho/llamafile/releases

The README doesn't really feel helpful in guiding me there.

I assumed the quickstart would jump straight into using llamafile the tool.

[Djip007]

If you want to run your "own" model, the simple for me is to download llamafile exec from release and get a gguf model. (no "install" is require, only make the llamafile executable)
then

llamfile -m "your_model.ggug" [....]

to use it.

(https://github.com/Mozilla-Ocho/llamafile?tab=readme-ov-file#running-llamafile-with-models-downloaded-by-third-party-applications)
llamafile-0.8.12 => is the simple executable.
llamafile-0.8.12.zip => you get bench exec, converter ... if you want to play some more.

[invisiblepancake]

learning? =)