Hello! We are working with API 3.5 and we need to at least roughly plan the ratio of input and output token consumption. But I can’t even imagine how much it is. According to my calculations, this is approximately 1 to 2 (input-output). I understand that the context is different, the tasks are different, but perhaps there is some known average number? Thank you!
That is something completely determined by your application.
“Read this short story, and give me a one sentence summary for my web site”
= low output to input ratio
Chatbot remembers everything someone ever said, along with all the AI responses:
= low output to input ratio
“Give 50 great cookie recipes”
= high output to input ratio
“Write complete OpenAI API documentation”
= high output to input ratio
1 Like
Very useful, thanks! * Text for 25 characters min. limit 