Mastodon Feed
cwebber@social.coop ("Christine Lemmer-Webber") wrote:
> Because of this I got a motherboard with slow GPU interconnect. It’s good for running many small experiments in parallel (which is my main use case) but horrible for any models split across gpus.
You spent $50k and your thing can't even actually make proper use of the GPUs you spent an approximate average US person's salary on?