Vector Database

What it is

A vector database stores and searches embeddings—numerical representations of text or other content. When you need to find relevant documents for RAG, you convert the query to an embedding and search for similar embeddings in the database. Examples include Pinecone, Weaviate, and Chroma.

Why it matters

Vector databases enable semantic search at scale. Traditional keyword search misses conceptual matches; vector search finds semantically similar content. If you're building RAG applications, you'll likely use a vector database to store and retrieve your document embeddings.