logoalt Hacker News

maz1byesterday at 7:57 PM2 repliesview on HN

AFAIK, they don't have any deals or partnerships with Groq or Cerebras or any of those kinds of companies.. so how did they do this?


Replies

tcdentyesterday at 8:01 PM

Inference is run on shared hardware already, so they're not giving you the full bandwidth of the system by default. This most likely just allocates more resources to your request.

show 1 reply
hendersoonyesterday at 7:58 PM

Could well be running on Google TPUs.