azure-ai-documentintelligence

What it is and what it does

This is the official Python SDK for Azure AI Document Intelligence, Microsoft's cloud-based document analysis service. It provides a client to submit documents for analysis using either prebuilt models (for receipts, invoices, business cards, ID documents, W-2 forms, and other common types) or custom models you train on your own data. The service extracts layout, text, key-value pairs, tables, barcodes, QR codes, formulas, and font/style information depending on the model and add-on capabilities you enable.

The package wraps the Azure REST API and handles authentication (API key or Azure Active Directory), request marshaling, and response parsing. It depends on azure-core for common Azure SDK patterns and isodate for date handling. The library supports both synchronous and asynchronous operations, though async requires aiohttp to be installed separately. It is actively maintained, production-stable, and widely used in the top 5000 PyPI packages by download volume.

Use it for:

• Extract structured data from receipts and invoices for expense tracking or accounting automation.
• Build a custom model to extract domain-specific fields from contracts, medical forms, or insurance documents.
• Classify incoming documents by type (e.g., invoice vs. receipt vs. purchase order) before routing to different processing pipelines.
• Read and extract text and layout from scanned PDFs or images as part of a document digitization workflow.
• Detect and extract barcodes or QR codes from documents using add-on capabilities.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Python client for Azure AI Document Intelligence, a cloud service that uses machine learning to extract text, structured data, and field values from documents via prebuilt and custom models.

Yes. This is a production-stable, actively maintained official Azure SDK with low install friction, permissive licensing, no known vulnerabilities, and a clear use case for anyone building document analysis into an Azure-backed application. Install it if you need to integrate Document Intelligence into a Python project; skip it if you don't use Azure or need a different document processing backend.

Install

azure-ai-documentintelligence on PyPI

pip install azure-ai-documentintelligence

uv add azure-ai-documentintelligence

poetry add azure-ai-documentintelligence

Installing azure-ai-documentintelligence

Before you install

Low friction: pure Python wheel with only three runtime dependencies (isodate, azure-core, typing-extensions). Actively maintained with recent releases; repository shows 5588 stars and last commit on 2026-08-14. No known vulnerabilities.

License in practice

MIT License (permissive): you may use, modify, and distribute this package freely in commercial and private projects with minimal restrictions, provided you retain the license notice.

Quickstart

pip install azure-ai-documentintelligence

from azure.core.credentials import AzureKeyCredential
from azure.ai.documentintelligence import DocumentIntelligenceClient

endpoint = "https://<resource>.cognitiveservices.azure.com/"
credential = AzureKeyCredential("<api_key>")
client = DocumentIntelligenceClient(endpoint, credential)

Requires Python 3.8 or later and an active Azure subscription with a Document Intelligence resource provisioned; async operations require aiohttp installed separately.

Verify before relying

• Whether the package's async transport (aiohttp) is commonly needed for typical use cases or mainly for specialized workloads.
• Performance characteristics when processing large batches or high-resolution documents with add-on capabilities enabled.

Package facts

Evidence: azure_ai_documentintelligence-1.0.2-py3-none-any.whl

Keywords: azure, azure sdk

Tags

More Artificial Intelligence packages

Further reading