Update llama.cpp quantize path

[gfvvz]

Update llama.cpp quantize path.

[shimmyshimmer]

It doesn't work as it's a directory
@Erland366 tried running./llama.cpp/examples/quantize and it doesn't work, whereas the original solution of ./llama.cpp/llama-quantize works

[gfvvz]

It doesn't work as it's a directory @Erland366 tried running./llama.cpp/examples/quantize and it doesn't work, whereas the original solution of ./llama.cpp/llama-quantize works

Oh, but on my side without this change, it report error. I think this may not the binary location, just quantize folder location, isn't it?

[danielhanchen]

I'm going to make it easier to use quantize directly - but it should be a binary

[gfvvz]

I'm going to make it easier to use quantize directly - but it should be a binary

I think that's a better idea.