AI-103: Developing AI Apps and Agents on Azure

An embedding model is designed to convert text (documents and queries) into numerical vector representations that capture semantic meaning. These vectors enable similarity-based search, such as vector search or retrieval-augmented generation (RAG), by allowing the system to compare the vector of a user query against the vectors of indexed documents to find the most relevant matches. The Azure AI Foundry (Microsoft Foundry) model catalog includes embedding models such as text-embedding-ada-002 and other embedding models specifically for this purpose, distinct from LLMs or SLMs which are optimized for text generation rather than producing embeddings.