This looks very interesting as majority of the world is moving towards bigger "do everything for you" LLMs. Just few preliminary questions after I glanced over

This looks very interesting as majority of the world is moving towards bigger "do everything for you" LLMs. Just few preliminary questions after I glanced over the repo and blogs: - there are also many small models that try to "do everything" like phi, mistral, etc., do you find llmware have better quality and performance? - how does the rag relate features compare with other tools like llamaindex?