It isn't clear to me the default chunking and tokenization that is being performed under VectorStoreIndex.from_documents(). Usually I can figure this out on my own, but having difficulty. Is this documented somewhere?

Quick question, is there a working example of using llamaindex to build a chat app that can search both a vector store and the internet to synthesize answers?

@Logan M Couple of quick questions for you. I looked for an example of combining neo4 and KnowledgeGraphIndex with HierarchicalNodeParser to extract hierarchical nodes from markdown documents containing headers, tables, etc. I couldn't find an example, and after some trial and error, I was unable to achieve this.

1. Does hierarchical node processor extract nodes based on headers and other document structure?
2. Is there a way to accomplish the above, or something like it?

I'm going off the recommendation that since I have legal-like structured documents, that extracting nodes and keeping the structure I can get the better than average results from a Q&A RAG app. I also, am assuming I can use neo4j since it is an established knowledge graph for better performance than simply storing the nodes on disk and querying from there.

I'm trying to come up with a set of steps to deploy an inference model on GPU in Azure that can scale to zero when not in use and spin-up when the endpoint is called.

This seems like it would be a common problem, because there are many companies that want a closed-off ChatGPT with RAG to prevent data leakage. Additionally, GPUs are expensive, so it is ideal to pay for only the compute that is used. I am assuming that Kubernetes is the best approach, but Kubernetes is not easy to work with directly for many reasons. I would therefore, expect that there is an existing framework or solution that makes this process easy. Are there some simple solutions to this problem?