Sinkhorn: Make LLMs even smaller through quantisation while maintaining accuracy | Hacker News

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

		Sinkhorn: Make LLMs even smaller through quantisation while maintaining accuracy (github.com/huawei-csl)
		4 points by ilitirit 87 days ago \| hide \| past \| favorite \| 1 comment

albertwang 86 days ago [–]

Media commentary: https://venturebeat.com/ai/huaweis-new-open-source-technique...

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