When it comes to indexing and embedding in the construction of BOT/RAG-like architectures for subsequent querying, it is crucial to know which algorithms or functions are used for data chopping, indexing, and embedding. This is essential for ensuring quick access to data or for providing precise and accurate responses.
Does Llamaindex use its own unique algorithms for these processes, or does it rely on third-party libraries, such as those from OpenAI? I'm seeking clarity on whether this is a generic approach in the industry and how critical it is in building these systems. Any insights or experiences shared by members in this area would be greatly appreciated.
