I don’t understand how to decide between doing slow or fast premium requests im premium account. My requests always counted as “fast premium request” in my profile request counter.
Can you explain how to do slow requests?
Can you explain how to do fast requests?
I used it for a trial period before premium. Which model was chosen during the trial?
Hey, you can’t switch between fast and slow requests. Slow requests start when your quota is exhausted. Your slow requests can be as fast as the fast ones, so you won’t have to wait for hours.
To see which models you’ve used, go to your account, click the toggle as shown in the screenshot, and your model usage logs will appear.
1 Like
Thanks for the screenshot! I’m a little confused- you say slow ones don’t start until quota is exhausted, but what governs fast requests of “Premium models” and gpt-40-mini or cursor-small, which here are labeled “fast” as well. You have a lot of quota left, yet it is using both sets of models. Do you know how that works? And can you control which one it uses if you are running close to your quota?
Hey, so all non-premium requests are always fast, meaning they are always responded to instantly. With a Pro subscription, you always have unlimited fast usage of non-premium models.
Only premium models have the “slow” tier, which happens when you run out of fast request allowance for those models.