Vector Database
A database that finds things by similar meaning instead of exact words.
In everyday terms
Store the embeddings of all your documents, then ask "what's closest to this question?" and it returns the most relevant passages.
For professionals
Indexes high-dimensional vectors for fast approximate nearest-neighbour search (e.g. HNSW), often with metadata filtering.
Think of it like…
A librarian who shelves books by topic-closeness, not alphabetically.
You've already seen it
Behind "chat with your documents" features.
Myth vs reality
Myth: You need a vector database to use AI.
Reality: Only for searching large collections by meaning. Many uses don't need one.
Quick check
A vector database is mainly used to…
- Store passwords
- Find content with similar meaning
- Train models
- Render images
Show answer
Find content with similar meaning: Similarity search over embeddings.