Vector Index
A data structure that makes finding the nearest vectors fast, even among millions of them.
Think of it like
A library’s shelving system — you don’t walk past every book, the organization takes you near the right shelf instantly.
Example
With ten million embeddings, a vector index returns the closest matches in milliseconds instead of scanning them all.
How it actually works
Comparing a query against every stored vector is fine for thousands but hopeless for millions, so a vector index organizes the vectors to skip most comparisons. The catch is that fast indexes are usually approximate — they trade a small chance of missing the true nearest neighbor for enormous speed. Tuning that accuracy-versus-speed dial, and the memory the index consumes, is the core operational concern of any vector system.
For product teams
The index is what keeps search fast as your data grows — and the accuracy-versus-speed tradeoff behind it is a real choice.
For engineers
A structure (e.g. HNSW, IVF) enabling sublinear approximate nearest-neighbor search over a vector set.
Related
- Similarity Search — The operation it accelerates.
- Vector database — The store it lives inside.
- Embeddings — What it organizes.
Read anything AI without the jargon
Look up any term in plain English, or save terms as you read with the free Chrome extension.
Open DecoderAdd to Chrome