#you are supposed to do that on your

1 messages · Page 1 of 1 (latest)

small zinc
#

I was going to track that on the backend but I see your point

austere latch
#

with the token tracking you are also supposed to manage and queue your clients requests so you wont get rate-limited by openai

small zinc
#

Not sure how to do that, just figured requests would throttle if they actually hit the rate limits

austere latch
#

if you dont add any mechanisms then it will lead to errors it there is too much usage at the same time