Llama index keeps refining over and over

Not sure what you mean. Can you share some code?

Figured it out. Well.. mostly

It's because of response mode settings https://docs.llamaindex.ai/en/stable/module_guides/deploying/query_engine/response_modes.html

When it's default, it refines something like 7-8 times so I have set it to simple_summarize

is there a way I can set it to refine just once?

I'm not sure what you mean 🤔 If it refines as much as it needs to read all retrieved context

To reduce the number of refines (if using a vector index), I would decrease the top-k

Ah I see. Thank you 🙏