Vector Database
A database optimised for storing and querying high-dimensional embedding vectors via similarity search.
Full Definition
A vector database stores high-dimensional embedding vectors alongside their source content and metadata, and supports efficient approximate nearest-neighbour (ANN) search — returning the k most similar vectors to a query vector in milliseconds even over millions of entries. Popular vector databases include Pinecone, Weaviate, Qdrant, Milvus, and pgvector (Postgres extension). They are the infrastructure layer for RAG systems: documents are chunked, embedded, and stored; at query time, the user question is embedded and the most similar document chunks are retrieved. Vector databases support multiple distance metrics (cosine similarity, dot product, L2 distance) and filtering by metadata.
Examples
Storing embeddings for 500,000 support tickets in Pinecone; a new ticket query retrieves the top-5 most similar past tickets for context.
A recommendation engine storing product embeddings in Qdrant and returning the 10 most similar items to a user's browsing history vector.
Apply this in your prompts
PromptITIN automatically uses techniques like Vector Database to build better prompts for you.
Related Terms
Embedding
A dense numerical vector that represents a token, sentence, or document in a con…
View →RAG (Retrieval-Augmented Generation)
Augmenting model responses by retrieving relevant documents from an external kno…
View →Grounding
Connecting model outputs to verifiable external sources to reduce hallucination …
View →