Léxico / CONCEPTO
Embeddings
Dense numerical vectors that represent text (or other data) in a way that captures semantic meaning, enabling similarity search and retrieval.
Embeddings convert text into arrays of numbers (vectors) where similar meanings land close together in vector space. “Dog” and “puppy” would have nearby vectors, while “dog” and “spreadsheet” would be far apart.
Common Uses
- Semantic search — find documents by meaning, not just keywords
- RAG pipelines — retrieve relevant context before generating a response
- Clustering — group similar content automatically
How They Work
- Text goes into an embedding model (e.g., OpenAI’s
text-embedding-3-small) - The model outputs a fixed-size vector (e.g., 1536 dimensions)
- Vectors are stored in a database and compared using cosine similarity