Not
Hacker
News
!
Home
Hiring
Products
Discussion
Q&A
Users
Sinkhorn: Make LLMs even smaller through quantisation while maintaining accuracy | Not Hacker News!