Amazon OpenSearch Service introduces Neural Search
Amazon OpenSearch Service now supports Neural Search on OpenSearch 2.9, enabling builders to create and operationalize semantic search applications with reduced undifferentiated heavy-lifting. For years, customers have been building semantic search applications on OpenSearch k-NN, but they’ve been burdened with building middleware to integrate text embedding models into search and ingest pipelines. Amazon OpenSearch Service customers can power Neural Search through integrations with Amazon SageMaker and Amazon Bedrock enabling semantic search pipelines that run on-cluster.
Amazon OpenSearch Service now supports Neural Search on OpenSearch 2.9, enabling builders to create and operationalize semantic search applications with reduced undifferentiated heavy-lifting. For years, customers have been building semantic search applications on OpenSearch k-NN, but they’ve been burdened with building middleware to integrate text embedding models into search and ingest pipelines. Amazon OpenSearch Service customers can power Neural Search through integrations with Amazon SageMaker and Amazon Bedrock enabling semantic search pipelines that run on-cluster.
With Neural Search, customers can use OpenSearch APIs to run search queries via human language and deliver more relevant results using text embeddings that incorporate semantic understanding and similarity. Customers can integrate text embeddings hosted on Amazon SageMaker or managed by Amazon Bedrock. Customers are empowered to quickly integrate state-of-the-art models and effectively manage models using Amazon SageMaker and Amazon Bedrock.
Neural Search is available in all AWS Regions that support OpenSearch 2.9+ on the Amazon OpenSearch Service. To get started, refer to the documentation.