Knowledge Bases for Amazon Bedrock now supports MongoDB Atlas for vector storage
Knowledge Bases for Amazon Bedrock securely connects foundation models (FMs) to internal company data sources for Retrieval Augmented Generation (RAG), to deliver more relevant and accurate responses. Today, we are announcing vector storage support for MongoDB Atlas in Knowledge Bases (KB) for Amazon Bedrock.
Knowledge Bases for Amazon Bedrock securely connects foundation models (FMs) to internal company data sources for Retrieval Augmented Generation (RAG), to deliver more relevant and accurate responses. Today, we are announcing vector storage support for MongoDB Atlas in Knowledge Bases (KB) for Amazon Bedrock.
Knowledge Bases’ native integration with vector databases allows you to innovate and create unique vector search based experiences, mitigating the need to build custom data source integrations. Vector search allows you to generate deep and accurate insights and find specific information from a corpus of documents. With this launch, your MongoDB Atlas vector database can now take advantage of Knowledge Bases capabilities such as adding metadata to source data to retrieve a filtered list of relevant passages, customizing prompts, and configuring the number of retrieval results. You can connect your AWS account to MongoDB Atlas over the public internet as well as through AWS PrivateLink for added security.
This integration adds to the list of vector databases supported by Knowledge Bases, including, Amazon Aurora, Amazon OpenSearch Serverless, Pinecone, and Redis. You can also use this integration with KB’s Retrieve API and the RetrieveAndGenerate API. The MongoDB integration for Amazon Bedrock Knowledge Bases is now generally available in the US East (N. Virginia) and US West (Oregon) AWS Regions. To learn more, refer to the MongoDB integration feature blog and Knowledge Bases documentation. To get started, please visit MongoDB Atlas on AWS Marketplace and the Amazon Bedrock console.