About Slow Request

non-premium models (gemini-1.x, 4o-mini, cusror-small) always use fast requests, since there’s no limit

there’s extensive previous discussion on the forums about it: What are fast and slow uses? - #6 by testacode

also, anthropic servers were down earlier today so no claude requests would have gone through

if you’re having trouble with models from anthropic, you can check their status page to see if servers are down