WARNING utils.py L37: For better inference performance, please install exllamav2 kernel

[Almenon]

This warning appears during AI initialization when running python main.py. Better inference performance would be nice because inference is pretty slow on old hardware.

WARNING utils.py L37: For better inference performance, please install exllamav2 kernel via pip install git+https://github.com/AutoGPTQ/AutoGPTQ.git@b8b4127

[Almenon]

I looked into installing AutoGPTQ, but found out that it is deprecated. It looks like autoround should be using GPTQModel instead. I raised a issue on their Github: intel/auto-round#428