I have an important inquiry. I have developed an application and integrated it with ChatGPT, and I plan to launch advertising campaigns for the application as I believe it will gain significant popularity. My question is about the API limit set by ChatGPT. I want to avoid encountering a high rate of requests from the users of the application and reaching the allowed limit for answering questions, which could result in the disruption of the AI service integrated into my application. Therefore, I would like to inquire about the existing limit, if any, and whether there is a mechanism to increase this limit. If there is such a mechanism, what are the associated costs, how can I request it, and what is the corresponding limit for each price tier?
Welcome to the community.
Here’s the docs for rate-limits for the API…
There’s also a billing limit that you set in your account. If you’re expecting a lot, I would set it as high as possible to start. Changing the billing cap for the month can sometimes take a bit of time.
The ChatGPT API has different rate limits based on subscription plans. Free trial users have 20 requests per minute (RPM) and 40,000 tokens per minute (TPM). Pay-as-you-go users have 60 RPM and 60,000 TPM for the first 48 hours, then 3,500 RPM and 90,000 TPM after. To increase limits, request a higher tier plan: Free (20 RPM, 40,000 TPM), Starter ($100/mo, 100 RPM, 200,000 TPM), Standard ($500/mo, 500 RPM, 1,000,000 TPM), Pro ($1,000/mo, 1,000 RPM, 2,000,000 TPM). Contact support to request an upgrade. Tips to avoid reaching limits: use max_tokens parameter, wait for token reset, and pause app when needed. Ensure a great user experience.