> ## Documentation Index
> Fetch the complete documentation index at: https://wiki.platelunchcollective.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Approximate Nearest Neighbor (ANN) Search

> Approximate nearest neighbor search is the algorithm that finds the vectors closest to a query vector in a large index.

*Technical implementation* · *AI Search Infrastructure*

## Definition

Approximate nearest neighbor search is the algorithm that finds the vectors closest to a query vector in a large index. It trades a small, controlled amount of recall for dramatic speed gains over exact search.

## Why It Matters for AI Search

Exact search over millions of vectors is computationally prohibitive at production scale. ANN algorithms — HNSW, IVF, and others — make vector retrieval fast enough for real-time query responses. The approximation means that a small percentage of genuinely relevant chunks may not appear in first-pass results. This is why the reranking stage exists: to catch precision failures that the approximate first pass missed.

## Related Terms

<CardGroup cols={2}>
  <Card title="First-pass retrieval" href="/ai-search-glossary/first-pass-retrieval" />

  <Card title="Reranking" href="/ai-search-glossary/reranking" />

  <Card title="Vector database" href="/ai-search-glossary/vector-database" />

  <Card title="Embedding" href="/ai-search-glossary/embedding" />

  <Card title="Dense retrieval" href="/ai-search-glossary/dense-retrieval" />
</CardGroup>

## Relevant Plate Lunch Collective Services

[AI SEO](https://www.platelunchcollective.com/services/ai-seo)