← Back to glossary
E

AI CONCEPTS

Embedding

Also known as: Vector Embedding

Definition

An embedding is a mathematical representation — typically a high-dimensional vector — of text, images, or other data that captures semantic meaning. Similar items have embeddings that are close together in vector space.

Simple Explanation

Imagine all sentences in the world plotted on a map. Sentences about cats cluster near other cat sentences, far from car-repair sentences. An embedding is the coordinates on this map. 'Best cat food' and 'What do kittens eat' would land very close.

Real-World Use Case

A semantic search engine converts all your documents to embeddings. When you search 'how to reduce costs,' it finds documents whose embeddings are closest — even if they don't use those exact words.

Examples

OpenAI text-embedding-3Cohere EmbedSentence-BERTBGE

Related Terms

Learn More

  • OpenAI Embeddings Guide
  • Pinecone Embeddings 101