Knowledge Maxxing← Back to topic library
AI MaxxingIntermediate75 sec
Vector database
A vector database stores embeddings and can quickly find the ones most similar to a given query vector, even across millions of stored items.
Prerequisite: Embedding
What matters
Standard databases are not built to efficiently compare high-dimensional vectors; a vector database uses specialized indexing to make similarity search fast at scale.
Example
A vector database storing embeddings of every support article can return the five most semantically similar articles to a new question in milliseconds.
Common misconception
A vector database is not simply a database with an extra column; it uses different indexing techniques designed specifically for similarity search over vectors.
Recall question
Why can't a standard database efficiently do what a vector database does?
Keywords and related concepts
vector databasesimilarity searchindexingembedding