I'm just about to create my very first plugin for ChatGPT and one thing I can't find an answer to is how rate limits work with plugins.
I know my own account has a 25 requests per hour, 200 RPD limit for GPT-4. If I'm using my API key in a plug that I release for general use, and ... over 200 people use it in a day (yeah, sure!), what will happen? Does the plugin continue to use my personal API key and crash after the 25th person uses it within the first 3 hour time period? Does OpenAPI instead look for the logged-in user and use *their* rate limit?
Is there a way to make a plugin use GPT 3.5 instead?
Any insights would be much appreciated!