What are you having good luck with on Apple Silicon? Or is this more of a statement for when local AI becomes “good enough?”

(FWIW I am with you; I haven’t found a local model that works well enough to be a daily driver)

qwen models are basically <opus and >sonnet. 397b runs at Q8 on m3 ultra. for mbp m5 max I'd use the +120b qwen model.