Hey What would you recommend as best way

Are you using the latest llama_index and langchain versions? There are some ChatGPT specific improvements in both. There's also a small demo here: https://github.com/jerryjliu/gpt_index/blob/main/examples/vector_indices/SimpleIndexDemo-ChatGPT.ipynb

If you are still encountering issues, you might need to increase similarity_top_k in your query call

so increasing similarity_top_k loads more than one context found from embeddings?

yeah I updated to 0.4.22 today, will check langchain also

I am experimenting with similar now. One thing I didnt expect was, @dagthomas shared a post yesterday with chunk size limit set to 600, and while I need to test a lot more, the smaller token size has been the most impactful parameter so far. I havent tried knowledge graph yet although I am about to try that one. The playground code looks especially useful for testing different combinations: https://github.com/jerryjliu/gpt_index/blob/main/examples/playground/PlaygroundDemo.ipynb

Keep in mind, smaller chunk size can be more expensive as, embeddings have a fixed size output. So if you stuff 4k tokens in an embedding versus 600, you need more embeddings per quantity of input data and accordingly more vectors to store.

Hey that playground code looks very useful to trying to figure this out, thanks mate

Exactly! By default, it's set to one

@adrianlee2220 I have been eager to try out the playground myself but havent had a chance yet and it may be a few days before I can, if you get the chance to share your experience or any pointers, I would be grateful!