3 ms·
I had an occasion recently where I was working a lot of late nights/early mornings with AI use. And I'd be getting these instant, beautiful responses, and then,
by jrumbut 6mo ago
I had an occasion recently where I was working a lot of late nights/early mornings with AI use. And I'd be getting these instant, beautiful responses, and then, as soon as the sun started coming in the windows, it would take longer and fail more, and by the time the clock struck 9 AM, every LLM had turned back into a pumpkin.
- jannes 6mo agoAre you saying US data centers idle in the night rather than serving European/Asian users?
- zemo 6mo agoideally european/asian users would hit european/asian servers, so potentially not surprising
- user_7832 6mo agoWhich service(s) were you using, if you don't mind sharing? I'm curious if most of the big players including eg Google do this thing of nerfing models or it's limited to more "smart" (read: black box models like ChatGPT.