Retrieval-Augmented Generation architecture on GCP with private network connectivity. Features Cloud Run for the RAG API, Vertex AI for embeddings and LLM inference, Cloud SQL for vector storage, Cloud Storage for document ingestion, and VPC Service Controls for network isolation. Fork this diagram on Diagrams.so to customize the vector database or add additional document sources for your RAG pipeline. Source: https://cloud.google.com/blog/topics/developers-practitioners