# What are embeddings?

Embeddings are numeric representations of text, images, or other data where similar meaning produces nearby vectors. They power semantic search, retrieval, clustering, and deduplication. An embedding model turns each document into a vector once; comparing vectors is then fast enough to search millions of items in milliseconds.

Embeddings are the retrieval half of RAG and the backbone of internal search, recommendation, and anomaly detection over business data.

They are cheap: embedding a million typical documents costs a few dollars, which is why every serious data boundary embeds everything on intake.

## Related terms

- [Inference](https://allocate.network/glossary/inference.md)
- [Tokens](https://allocate.network/glossary/tokens.md)
- [Retrieval-augmented generation](https://allocate.network/glossary/rag.md)

---

[HTML page](https://allocate.network/glossary/embeddings) · [Machine-readable catalog](https://allocate.network/catalog.json)