I still did not get my answer. If anyone has any experience I would appreciate help. The problem is:
We can store the entire text conversation for a chatbot, how can we calculate the total pricing for that conversation, by counting the number of input tokens and output tokens, given that for every user query, the entire previous conversation plus the system message is the input to the model? I am asking this question because, for the chatbot that I have, I want to measure the pricing per user.