In my billing account, there is a chart. That chart consists of prices for API calls also. Does anyone have any idea about is open AI charges for API calls, if yes then in assistant API we have to check the run status again and again, then what about charges? @logankilpatrick
Hi,
There is no fixed price per call for the assistants API if you are using retrievals, the engine behind the scenes will make real-time decisions as to the information required to best answer the prompt, this could be a vector search, a full file load into context or a combination of both, the assistants API is not production ready currently and is there to be evaluated, it is actively being worked on and there will be updates to the system from time to time to try new ideas and get more feedback.
thanks for your reply, I want to know that in my code I am creating an assistant with retrievals and code interpretation. After that i created a thread and then run. after run i need to check run status again and again. So my question is, Does open ai charge for checking status also. Because i need to check it 50 to 60 times for each prompt.
If you are performing 50-60 check then I would increase the time between checks by a factor of 10 to reduce that down to just a few checks. Checks do not count as tokens used, but 50-60 checks is excessive and will put unnecessary load on the system.
Thanks a lot for your support. i will surely increase the interval