Model quality relative to other quantization techniques?

#1
by spanspek - opened

I've heard good things about these quants but can anyone tell me how they perform relative to say unsloth?

This is a big download for the country I'm in so I can't easily afford to just download several large models and test for myself

I'm trying out a small model first but having problems: https://huggingface.co/Intel/Ling-flash-2.0-gguf-q2ks-mixed-AutoRound/discussions/1

If I can get it working then I'll try the GLM AutoRound gguf.

Sign up or log in to comment