Vector database
A vector database is a specialised database that efficiently stores vector embeddings and allows similarity search, essential for RAG systems.
Classic databases (PostgreSQL, MySQL) aren't optimised for vector-similarity search over hundreds of millions of vectors. Vector databases (Pinecone, Weaviate, Qdrant, Chroma, Milvus, pgvector for PostgreSQL) use algorithms like HNSW or IVF for sub-second similarity retrieval at scale.
Example
A company builds an internal 'ask your docs' chatbot. 50,000 documents are embedded and stored in Qdrant. Query → embedding → top-5 most similar chunks from vectorDB → fed as context to LLM → grounded answer.
Frequently asked questions
Which vector DB should I pick?
Budget & simplicity: pgvector (existing PostgreSQL). Managed & scale: Pinecone. Open source & control: Qdrant or Weaviate. Embedded/local: Chroma.
Related terms
Further reading
- → Our service: AI sector