An embedding model converts a sentence, paragraph, or document into a list of numbers. Texts with similar meanings produce vectors that are close to each other, even when they share no words. That makes embeddings the foundation of semantic search, recommendation, clustering, and the retrieval step in RAG systems.
Embeddings are usually stored in a vector database or a database extension such as pgvector. Choices about which model to use, how to split documents, and how to combine semantic and keyword search have a large effect on retrieval quality.