It's not quantising existing models, they're training new ones. | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		IanCal on Feb 28, 2024 \| parent \| context \| favorite \| on: The Era of 1-bit LLMs: ternary parameters for cost... It's not quantising existing models, they're training new ones.

leroman on Feb 28, 2024 [–]

I understand this part but it seemed that the 16->8->4 etc is similar to compression of the "net" and seemed to lower quality below 8.

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