Amazon S3 Vectors now supports hybrid search combining vector and full text for more relevant search results
News
This article announces hybrid search support in Amazon S3 Vectors, combining vector similarity search with full-text search in a single query for improved relevance.
- Hybrid search combines semantic meaning and exact term matching for better RAG and search results
- Maintains subsecond latency for infrequent queries and ~100ms for frequent queries
- Reduces vector storage and query costs by up to 90% compared to alternatives
- Uses BM25 relevance scoring with language-agnostic or language-specific text analyzers
- Employs Reciprocal Rank Fusion to merge vector and full-text results into single ranked list
- Available in all commercial AWS Regions and AWS China Regions
Hybrid search enables customers to store searchable text alongside vectors and perform combined queries through existing S3 Vectors APIs.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
Related articles
Sep 30
2026
2026
Amazon S3 Vectors now supports metadata pre-filtering for higher recall on filtered searches
Sep 30
2026
2026
Amazon S3 Vectors introduces metadata pre-filtering for up to 5x higher recall on filtered search
Jun 16
2026
2026
Amazon S3 Vectors now supports up to 10,000 similarity search results per query
Nov 10
2026
2026
Amazon ElastiCache Serverless now supports vector, full-text and hybrid searches plus aggregations
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.