Alpha zero used 5000 TPUs to generate games (inference only), and *16* to train ... | Hacker News

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

		gpm on Aug 30, 2022 \| parent \| context \| favorite \| on: Are GPUs Worth It for ML? Alpha zero used 5000 TPUs to generate games (inference only), and 16 to train the networks. The split definitely depends on what you're doing past developing/deploying. (Source: https://kstatic.googleusercontent.com/files/2f51b2a749a284c2...)

varunkmohan on Aug 30, 2022 [–]

Completely agreed. For some of these large language models, it would take a long time before inference spend dominates training spend.

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