Stuck at "Create the fine-tuning datasets.."

I’m stuck at “Create the fine-tuning datasets for Q&A and discriminator models.” in the olympics 3 documentation.
It uses the deprecated API endpoint, search.

Does the olympics 3 documentation need to be updated? The " To answer questions based on text documents, we recommend the procedure in Question Answering using Embeddings" is confusing me. Is it a followup or a replacement for the olympics 3 documentation?

I would like to fine tune model specialized for Q&A chat bot. I have done the olympics 1 & 2. I need technical support.