This repository is confusing, is it for windows?

[Njasa2k]

from "The main goal is to run the model using 4-bit quantization on a MacBook" means I'm guessing it's for iOS and I have no business being in this repo.

There's a mix of windows and apple commands and I can't tell which.
`# obtain the original LLaMA model weights and place them in ./models

ls ./models
65B 30B 13B 7B tokenizer_checklist.chk tokenizer.model`

Does the ls command even work on windows?

[x02Sylvie]

Instructions may be confusing but llama.cpp works both on windows, linux and mac

[Njasa2k]

How do I know which instructions is for windows and which instructions is for mac? is there a straight forward instruction?
I got 8GB of VRAM and 64GB of RAM, Windows, NVIDA RTX 2070.

[BetaDoggo]

The Windows equivalent of the ls command would be dir, but that command is just to check that the model files are in the correct location.

The rest of the instructions in that section should still apply but with ./quantize replaced with quantize.exe, same with ./main and main.exe.

[Wolvensdale]

Im trying to follow but "make" command won't work. Do you have some step by step for windows? I tried install cmake, and all visual studio, node js, etc etc all already installed.

[BetaDoggo]

I believe the build commands for cmake are the following:

cmake .
cmake --build . --config Release

Then I think the binaries will be put in the ./bin/Release folder.