AWS announced the general availability of vector search in Amazon DynamoDB, enabling real-time indexing and approximate nearest neighbor queries with 99%+ recall and single-digit millisecond latency, even at trillions of vectors scale. This supports applications in AI memory retrieval, product similarity, recommendations, and semantic search by storing and querying vector embeddings alongside traditional attributes in a serverless environment.
DynamoDB vector search allows partition key selection for scaling and attribute filters to refine results. It integrates with models such as those on Amazon Bedrock and offers the usual DynamoDB benefits: zero infrastructure management and pay-per-use. AWS framed this as a breakthrough for handling large-scale vector datasets, although independent performance assessments were not provided.
Source & further reading
Amazon DynamoDB now supports real-time vector search ↗Source published 2026-08-05. This report was prepared with AI assistance from the linked primary-source material. Claims and limitations should be checked against the original announcement or research.