rag-with-amazon-postgresql-using-pgvector-and-sagemaker and rag-with-amazon-opensearch-and-sagemaker
These are ecosystem siblings—both are reference implementations of RAG pipelines using SageMaker for embeddings and LLMs, but they demonstrate the pattern with different vector database backends (PostgreSQL with pgvector versus OpenSearch), allowing users to choose based on their existing infrastructure or requirements.
About rag-with-amazon-postgresql-using-pgvector-and-sagemaker
aws-samples/rag-with-amazon-postgresql-using-pgvector-and-sagemaker
Question Answering application with Large Language Models (LLMs) and Amazon Postgresql using pgvector
About rag-with-amazon-opensearch-and-sagemaker
aws-samples/rag-with-amazon-opensearch-and-sagemaker
Question Answering Generative AI application with Large Language Models (LLMs) and Amazon OpenSearch Service
Implements retrieval-augmented generation by storing document embeddings in OpenSearch and dynamically retrieving relevant passages to augment LLM prompts, addressing token limits and improving answer accuracy. Deploys SageMaker endpoints for both text generation and embedding creation, with infrastructure-as-code (CDK) for the full stack including OpenSearch clusters and credential management. Provides a complete end-to-end workflow from data ingestion through a Streamlit frontend, leveraging LangChain for orchestration.
Scores updated daily from GitHub, PyPI, and npm data. How scores work