I am using the completions API to try to create documents based on a lot of information. It seems like at the very end of my automated conversation that it’s exceeding the rate limit
Request too large for gpt-4-turbo-preview in organization org- on tokens per min (TPM): Limit 30000, Requested 36575
I looked up the rate limits here:
Based on what I spent, I would expect to be in tier 2. How can I verify this? I would expect to have a TPM of 450,000. I specified to use gpt-4-turbo. Is there a reason why it’s preview?
I am trying to use a model that can handle a decently sizeable token limit and be able to handle up to 100,000 TPM. Do you have any suggestions?