New top story on Hacker News: ML training compute has been doubling every 6 months since 2010

ML training compute has been doubling every 6 months since 2010
26 by innerfestival | 6 comments on Hacker News.