Azure Document Intelligence is a powerful AI capability within Azure Content Understanding in Foundry Tools (formerly Azure AI Services) that enables organizations to automatically extract text, key-value pairs, tables, and document structure from files such as PDFs, images, and forms. It uses advanced AI and optical character recognition (OCR) to transform unstructured content into structured, actionable data. As part of our Content Understanding capabilities, Azure Document Intelligence provides high-accuracy and reliable deterministic extraction from structured documents while Content Understanding also offers LLM-powered analyzers for complex, unstructured, and multimodal content. Together, they make it easier to prepare data for intelligent agents and applications that can read, analyze, and respond to real-world content with precision and speed.

Yes, Azure Document Intelligence—formerly part of Azure AI Services—is now rebranded as Document Intelligence in Foundry Tools. This change is part of a broader platform unification under Foundry, aligning these services with how developers build intelligent, agentic applications today.

Document Intelligence still offers the same powerful capabilities—like extracting text, tables, key-value pairs, and layout from documents, forms, and images using prebuilt and custom models. Document Intelligence is a part of our Content Understanding capabilities and a core tool within the Foundry ecosystem, designed to work seamlessly with other tools like Speech, Language, and Translator.

This rebrand helps clarify how Document Intelligence fits into the Foundry platform—along with Content Understanding’s advanced multimodal extraction capabilities, making it easier to discover, orchestrate, and integrate into modern AI workflows for document automation, knowledge extraction, and agentic reasoning.

See Azure Document Intelligence availability.

It supports printed and handwritten forms, PDFs, and images.

Document Intelligence supports multiple languages including English, French, German, Italian, Spanish, Portuguese, Dutch, Chinese, Japanese, and Korean. See the full list here.

The custom option uses five samples to learn the structure of your documents and intelligently extract text and data tailored to them. The prebuilt option doesn't use samples—the model has already been trained and is ready to use off the shelf to extract key fields from specific document types.

Layout extracts text using high-definition optical character recognition (OCR) tailored for documents. It also extracts the structures of tables (row and column numbers), selection marks and more.