Hi all,
I am using Open AI assistants API to develop the chatbot for my company, I need to bill each customers based on their usage, I am planning to leverage tokens they consumed per question and bill them accordingly.
I am confused on the pricing model here, since I am able to get the
- prompt_tokens ( I assume that as input tokens costing $0.00250 / 1K input tokens)
- completion_tokens (I assume that as the output tokens $0.01000 / 1K output tokens).
I am not sure how to get the cache_tokens (since it has a different billing rate) as the usage object in python does not fetch the cache tokens ( total tokens = prompt+completion).
Can anyone suggest the best billing strategies how to charge per user.