Amazon ElastiCache Serverless Adds Vector and Full-Text Search with Microsecond Latency

Amazon ElastiCache Serverless for Valkey now lets you run vector, full-text, and hybrid searches with aggregations—all with microsecond latency and up to 99% recall for vector queries. The feature rolls out across all commercial AWS Regions, AWS GovCloud, and the European Sovereign Cloud, making it perfect for latency-sensitive apps, RAG systems, and AI agents needing instant context retrieval.