The paper introduces QMoE, a compression and execution framework that allows trillion-parameter language models to be run efficiently on affordable hardware with minimal accuracy loss.

https://arxiv.org/abs//2310.16795

YouTube: https://www.youtube.com/@ArxivPapers

TikTok: https://www.tiktok.com/@arxiv_papers

Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016

Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers