Hi everyone,
I’m new to the OpenAI API and am trying to understand how Assistants work. If I uploaded 200 PDFs (knowledge base), and asked the Assistant a specific question about the content in the PDFs, would the number of input tokens be the entirety of the 200 PDFs (or rather a smaller subset of that based on what the assistant needs to answer that question)? Let’s say each PDF is 20,000 Input Tokens, would I be charged on 200 files x 20,000 tokens = 4,000,000 Input Tokens each time I ask a question? I also noticed the charge on cached Input Tokens is half the price, so if I ask a follow-up question, am I charged on only half? Thanks!