Better go for a less-quantized model even if it's slower than go for a faster, q... | Hacker News

Hacker News .hnnew | past | comments | ask | show | jobs | submit

		stavros 44 days ago \| parent \| context \| favorite \| on: I ran Gemma 4 as a local model in Codex CLI Better go for a less-quantized model even if it's slower than go for a faster, quantized one.

hmokiguess 44 days ago [–]

Thank you!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact