Azure Document Intelligence

API

Azure API for OCR, layout and prebuilt invoice, receipt and ID extraction

Price: Free tier, then paid
Access: Resource key and endpoint

About

Microsoft Azure service (formerly Form Recognizer) with Read and Layout models, prebuilt models for invoices, receipts, IDs, bank statements and tax forms, and trainable custom extractors. Analysis is asynchronous. Read and Layout also run in Docker containers on your own hardware.

What you can do with it

Extract text, tables, paragraphs and selection marks from PDFs and scans
Pull typed fields from invoices, receipts, IDs and pay stubs with prebuilt models
Train custom extraction and classification models on your own labeled documents

Get started

Create a Document Intelligence resource in the Azure portal (F0 is free)
Copy the resource endpoint and key
Poll the Operation-Location URL from the 202 response for results

Example

curl -i -X POST "$DOCUMENT_INTELLIGENCE_ENDPOINT/documentintelligence/documentModels/prebuilt-invoice:analyze?api-version=2024-11-30" \
  -H "Content-Type: application/json" \
  -H "Ocp-Apim-Subscription-Key: $DOCUMENT_INTELLIGENCE_KEY" \
  -d '{"urlSource": "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-REST-api-samples/master/curl/form-recognizer/sample-invoice.pdf"}'

Details

Hosting: Hosted service, Self-hosted
Available in: Worldwide
Official SDKs: C#, Java, JavaScript/TypeScript, Python
MCP server: None
Works with: Azure