You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is it possible to define rate limits at the granularity of the OpenAI API? I.e., different RPM/TPM for each model? Context is that we want to give 100 students access to the API at the same time through our tier 3 token so every key needs to have 35 RPM/1600 TPM for gpt-3.5-turbo and 50 RPM/5000 TPM for text-embedding-ada-002. If I understand the documentation correctly we could only set the minimum of each limit currently?
The text was updated successfully, but these errors were encountered:
The text was updated successfully, but these errors were encountered: