any plans to support hybrid search like

we'll add this support, but tbh in a very general sense all of llamaindex is hybrid search, we have a lot of advanced primitives that allow you to do a ton of stuff over your data (see use cases https://gpt-index.readthedocs.io/en/latest/use_cases/queries.html)

hybrid search purely in the form of keyword lookup + embeddings will be added soon

yes that’s what I meant, keyword to improve bad embeddings basically

Adding to this, just a question. I know pinecone and weaviate have their own sparse+dense vector hybrid search. Would you build these capabilities directly into llama index or utilize a pass through to their solution?

we just added this! https://twitter.com/gpt_index/status/1644487657505030146?s=20 cc @confused_skelly @Tiago Freitas

Holy geez man

Do you ever sleep! Haha. Mad respect for your productivity