undefined | Better HN

0 pointsbobxmax5mo ago0 comments

Who cares? Are you not seeing the exponential decrease in GPU inference costs? H100s costs 3x what they did a year ago.

0 comments

I'm not sure I understand how B (H100s costs 3x what they did a year ago) follows from A (exponential decrease in GPU inference costs).

j / k navigate · click thread line to collapse

0 pointsbobxmax5mo ago0 comments

Who cares? Are you not seeing the exponential decrease in GPU inference costs? H100s costs 3x what they did a year ago.

oblio5mo ago

I'm not sure I understand how B (H100s costs 3x what they did a year ago) follows from A (exponential decrease in GPU inference costs).

j / k navigate · click thread line to collapse