The issue is it’s cpu compute which is underutilized in gpu clusters anyway, so ... | Hacker News

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

boroboro4 22 days ago | parent | context | favorite | on: GigaToken: ~1000x faster Language model tokenizati...

The issue is it’s cpu compute which is underutilized in gpu clusters anyway, so practically it’s not really 1/1000.

pipsterwo 22 days ago [–]

Totally, edited my comment to specify "based on the assumption above." The main takeaway I was going for was 0.1% is not a small number in this context

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