Azure Document Intelligence is a Microsoft Azure service that uses AI models to extract text, key-value pairs, tables and structure from documents automatically. Prebuilt models and layout APIs read forms, receipts, invoices and cards without manual labeling by document type, and custom extraction trains on a few of your own documents for nonstandard layouts. The service is exposed through a REST API and is now part of Azure Content Understanding in Foundry Tools. Containers allow deployment to Azure Kubernetes Service, Azure Container Instances or a Kubernetes cluster on Azure Stack.
The service targets developers, data engineers and enterprise teams that handle large volumes of paperwork, such as finance, insurance, healthcare and logistics operations. Teams that need quick results use prebuilt models, while those with unusual forms train custom models and add human input to refine accuracy. Organizations with strict data residency rules can run the containers on-premises or at the edge, and enterprise security and privacy controls apply to both data and trained models. It also fits into the wider Azure stack alongside Azure AI Search, Azure OpenAI, Speech and Translator.
Typical use cases include automating claim, invoice and receipt processing by piping extracted fields into back-end systems, building intelligent search indexes so staff can find items such as total accounts payable across thousands of forms, and feeding extracted data into visualization tools for analysis. Developers wire the REST interface into business workflows that digitize paper forms, onboarding packets or contracts. Custom models help when documents deviate from traditional formats, and human feedback improves results over time. Free and standard pricing options let teams test before processing documents at scale.
