With hnsqlite I integrate these two powerhouses and add support for filtering metadata at embedding search time. While working through several LLM projects I found this to be very useful for self-hosted projects up to a reasonably large size (at least 5-10 million embeddings).