I’m creating an assistant with 1 data file (1MB) that I want to use. I’m wondering what is the cost of using it. In playground it shows great numbers for input. Even 20000 tokens. Is it the cost of every prompt via API?
Shouldn’t it be much cheaper?
1 Like
Heya! Welcome to the OpenAI Dev Community Forum!
Yes, Assistants can get expensive because you do not control what is included in “context”… So, most build out their own RAG system to have more control. It’s more work, but you have more power to get things done and control costs.
2 Likes
Thanks! What do You mean “RAG” ? How to do it?