~ pick your vector store based on constraints, not vibes ~
START HERE Already running Aurora PostgreSQL or have strong Postgres ops skill? YES NO Aurora pgvector SQL + vectors together. Cheaper than OpenSearch at low-to-mid scale. Primary use case is enterprise doc / FAQ search & you want fully-managed, no indexing? YES NO Amazon Kendra Managed enterprise search. 40+ built-in connectors. Pricey but zero ops. Need hybrid search (vector + keyword) and/or massive scale? YES NO OpenSearch Serverless Native hybrid search. Scales automatically. Default for Bedrock Knowledge Bases. Quick prototype or want to stay in one service? YES S3 Vectors cheap, simple, preview 3rd party (Pinecone, MongoDB) 💡 default: OpenSearch Serverless for most KB setups

Why these options

OpenSearch Serverless The default for Bedrock Knowledge Bases. Supports true hybrid search (BM25 + k-NN) natively in one query. Scales automatically without capacity planning. Higher cost floor than pgvector but pays off at mid-to-large scale.
Aurora PostgreSQL + pgvector Best if you already run Aurora for transactional data — you get to keep SQL joins between vectors and relational rows. Lower cost at small-to-medium scale. You run ops (index tuning, vacuum). No native hybrid search without extensions.
Amazon Kendra Not a raw vector store — a managed enterprise search service with built-in connectors to SharePoint, Confluence, Salesforce, S3, etc. Use when your corpus is scattered across SaaS tools and you want zero indexing work. The priciest option.
S3 Vectors Newest option (launched 2024). Stores vectors directly in S3 with k-NN search. Cheapest for read-light workloads. Good for prototypes, small corpora, or use cases where you want to stay in S3.
Exam angle When a stem mentions "hybrid search," "keyword + semantic," or "high scale" → OpenSearch. When it mentions "already running Aurora" or "cost-sensitive with relational data" → pgvector. When it mentions "SharePoint / Confluence / enterprise doc search" → Kendra. These keywords are the tell.

Related trees

Tree 2: Bedrock vs SageMaker deployment · Tree 3: Cost optimization · Tree 4: RAG troubleshooting
Architecture deep-dive: Pattern 1: Basic RAG · Pattern 2: Advanced RAG