AWS adds native vector search to DynamoDB with single-digit millisecond latency

Amazon DynamoDB now supports native vector search, enabling developers to store vector embeddings alongside operational data and perform similarity searches directly within the database. The feature delivers single-digit millisecond latency at 99%+ recall and is designed to scale to trillions of vectors without requiring separate infrastructure or management overhead.

Previously, adding vector search to DynamoDB required copying data into a dedicated vector database and maintaining a synchronization pipeline, which introduced operational overhead, data movement costs, and licensing expenses. DynamoDB’s native vector search eliminates these requirements by integrating the capability directly into the database service.

Vector indexes in DynamoDB have no storage limits and scale horizontally as data grows. The service also supports zero downtime maintenance, no server provisioning or patching, and no software installation or operation. These properties make it suitable for applications such as retrieval-augmented generation, recommendation engines, personalized experiences, anomaly detection, and agentic memory.

The announcement was made in an AWS News Blog post authored by Esra Kayabali and dated 05 AUG 2026.