What are the guidelines and limitations regarding the usage and management of secret keys for the ChatGPT API, particularly with respect to the gpt-3.5-turbo model, and what steps should be taken if a key is prematurely removed or to monitor usage against

Good morning, you would like to know certain things about the use and management of the ChatGPT Api secret keys.

I am currently using the gpt-3.5-turbo model and I have the following questions.

  1. How many questions can I ask for each secret key I generate.

  2. How should I renew the secret key once the usage limit has been exhausted? There is an endpoint to renew it.

  3. Yesterday I was testing and I had the following mishap: after creating the secret key and asking a single question, it was automatically removed from my created list. What should I do if that happens to me again?

  4. Is there a way to know what my consumption and my consumption limit are in real time? Is there an endpoint that consults that information?

  5. Is there a corporate or business plan that I can purchase?

Thank you for your attention and I look forward to your response.

Welcome to the forum!

Please create a shorter title. :slightly_smiling_face:


  1. You can send as many requests as your rate limits allow and as long as you are willing to pay for more requests. In turn that will help to increase your rate limits.

  2. The rate limits apply to your account not to a single key. You would deploy different keys for different apps or customers for example.

  3. I don’t know how you managed to automatically delete the key. Don’t press the delete button, I guess, or keep a couple of back-up keys around.

  4. You have to consult the usage tab in the account or create your own management system, which means routing all request through a dedicated gateway and managing consumption from there.

  5. Enterprise plans apparently come with better terms but you have to contact sales@openai.com and make a RFQ.

Hope this helps!