AWS claims 90% vector cost savings with S3 Vectors GA, calls it 'complementary' - analysts split on what it means for vector databases

Amazon S3 Vectors has been launched, integrating native vector storage and similarity search directly into AWS's object storage service. This development allows organizations to bypass the need for separate, specialized vector databases for operations like semantic search and RAG. The new service offers significantly increased capacity, scaling up to 2 billion vectors per index and 20 trillion per bucket. AWS claims S3 Vectors can reduce vector storage and querying costs by up to 90%. However, AWS positions S3 Vectors as a complementary solution, a performance tiering option, rather than a direct replacement for dedicated vector databases. Specialized vector database providers argue that their offerings still hold performance advantages for low-latency, high-throughput workloads. Analysts are divided on whether vector search will remain a standalone product category or become commoditized by cloud platforms. The decision for enterprises will depend on workload requirements, particularly concerning latency tolerance. For less time-sensitive operations and large-scale storage, S3 Vectors is presented as a cost-effective solution. Conversely, latency-sensitive use cases, such as real-time recommendations, will continue to necessitate specialized vector databases. A hybrid approach, combining both, is likely to be adopted by many organizations.