Amazon Announces New Updates to Amazon Bedrock

15 February, 2024

No Comments

AWS announced new updates to Amazon Bedrock. Amazon Bedrock is a new service that was announced at AWS re:Invent 2023 that allows organizations to provide information from their own private data sources to improve the relevance of answers.

Significant enhancements have been made since the launch of Amazon Bedrock, such as the introduction of Amazon Aurora PostgreSQL-Compatible Edition as an additional option for custom vector storage along with other options such as the vector engine for Amazon OpenSearch Serverless, Pinecone and Redis Enterprise Cloud.

One of the new updates is a richer selection of built-in models. In addition to Amazon Titan Text Embeddings, users can now choose from Cohere Embed English and Cohere Embed Multilingual models.

This update aims to provide users with greater flexibility and precision in how they manage and use their data in Amazon Bedrock.

In addition to these enhancements, users can now choose from a range of supported options, tailoring the backend to their specific requirements. Such features ensure that the integration of knowledge bases with existing data management systems is seamless and efficient, enhancing the overall utility of the service.

Many of Amazon Aurora’s database features will also apply to vector-embedded workloads, such as elastic storage scaling, low-latency global reads, and faster performance compared to open source PostgreSQL. Pinecone serverless is a new serverless version of Pinecone, which is a vector database for building generative AI applications. Thus, users will have more variety and scalability in choosing vector storage solutions.

Another important update to pay attention to is the existing Amazon OpenSearch Serverless integration. Redundant replicas are now disabled by default, which Amazon estimates will cut costs in half.

According to Antje Barth, chief developer advocate at AWS, Amazon Bedrock was created by Amazon with the goal of improving the user experience and offering flexible, cost-effective solutions for managing vector data in the cloud.