Slacker News Slacker News logo featuring a lazy sloth with a folded newspaper hat
  • top
  • new
  • show
  • ask
  • jobs
Library
← Back to context

Comment by YetAnotherNick

2 months ago

You can say it takes 1800-5400 W. Not sure where you are estimating it from.

2 comments

YetAnotherNick

Reply

KaiserPro  2 months ago

Depending on what you're doing its taking up to 8GPUs working in parallel to serve those queries.

  • YetAnotherNick  2 months ago

    Yes but then the batch size is in 100s or even 1000s. These GPU doesn't serve just 1 user at a time.

Slacker News

Product

  • API Reference
  • Hacker News RSS
  • Source on GitHub

Community

  • Support Ukraine
  • Equal Justice Initiative
  • GiveWell Charities