Hi everyone so I have built a program that uses Open AI API, I have a few people interested and realized that some users will be using my company’s (solo dev) API key for more requests than other users and I believe they should not be charged the same. I am thinking something like… ex: "basic plan (100 requests a month) " pro plan (1000 request a month) how do I figure out how much data one person uses? I need this to figure out how many API pulls someone can do a month to figure out my profitability, and what is a fair usage per user. If anyone has any ideas let me know. Thank you.
I might have a similar requirement if we decide to make our internal app publicly accessible. From what I understand, OpenAI’s pricing is based on the number of tokens used (both input and output).
Each plan you offer would need to define a maximum token usage limit. Additionally, you would need to track each user’s API calls and the corresponding token consumption. Depending on the user’s plan and remaining token quota, you could then determine whether to process the call or deny it.