Per the very short article, the solution was to pack multiple models per GPU. | Hacker News

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

		svachalek 87 days ago \| parent \| context \| favorite \| on: Alibaba Cloud says it cut Nvidia AI GPU use by 82%... Per the very short article, the solution was to pack multiple models per GPU.

make3 87 days ago [–]

yes but that could mean a layer per model

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