Agentic Document Extraction Python Library

The landingai-ade Python library provides a robust interface to the LandingAI Agentic Document Extraction REST API. It empowers developers to programmatically extract structured data from documents with flexibility, reliability, and performance, supporting complex document processing workflows.

How It Works

This SDK leverages schema-based extraction, allowing users to define data structures using Pydantic models which are converted to JSON schemas for the API. It offers both synchronous and asynchronous clients, with the latter supporting large document processing via asynchronous jobs. The library includes features like built-in retries with exponential backoff and pluggable HTTP backends (httpx, aiohttp) for enhanced flexibility.

Quick Start & Requirements

• Primary install / run command: pip install landingai-ade
• Non-default prerequisites and dependencies: Python 3.9+
• Links: Docs, Playground

Highlighted Details

• Fully-typed SDK with Pydantic response models for improved developer experience and error catching.
• Dual support for synchronous and asynchronous API interactions.
• Asynchronous job processing for handling large documents efficiently.
• Automatic retries with exponential backoff for network resilience.
• Schema-driven data extraction using Pydantic models.
• Pluggable HTTP client backends (httpx, aiohttp).

Maintenance & Community

The project maintains a community presence via Discord (https://discord.com/invite/RVcW3j9RgR) and publishes updates and insights on its blog (https://landing.ai/blog). Comprehensive documentation is available at https://docs.landing.ai/ade/ade-overview.

Licensing & Compatibility

The library's license information is available via its PyPI page. Specific compatibility notes for commercial use or closed-source linking are not detailed in the provided README snippet.

Limitations & Caveats

The README does not explicitly state limitations, alpha status, or known bugs. It does provide guidance on differentiating between null API response values and missing fields using Pydantic's model_fields_set.