Definition
Embeddings are the foundation of modern semantic search and RAGRetrieval Augmented Generation - AI architecture where the model is supplemented with documents retrieved in real-time to provide fact-based answers. systems. Text is converted to a vector of hundreds or thousands of numbers, where proximity in the mathematical space corresponds to semantic similarity. "Grid rent" and "tariff" will be close to each other in a well-trained library, even though the words differ. Embeddings are used in vector databases where you can search by meaning rather than exact keywords.