LandingAI Launches Agentic Document Extraction for Unlocking Insights from Complex Documents

Transforming Enterprises Complex Document Workflows and Understanding with Agentic Vision Technologies

PALO ALTO, Calif., May 28, 2025 /PRNewswire/ -- LandingAI, a pioneer in agentic vision technologies, today announced the major upgrades to Agentic Document Extraction (ADE). Unlike traditional optical character recognition (OCR), ADE sees a PDF or other document visually, and uses an iterative workflow to accurately extract a document's text, diagrams, charts, form fields, and so on to produce an LLM-ready output.

Document processing remains a significant hurdle for many enterprises, often incurring high costs and requiring substantial manual intervention. Agentic Document Extraction (ADE) goes beyond traditional OCR and fine-tuned LLMs by utilizing layout-aware parsing, visual grounding, and no-template setup, allowing for quick deployment and dependable outcomes without the need for fine-tuning or model training.

ADE addresses common challenges in document extraction and has made major improvements in speed and accuracy since the initial release:

• Fast: ADE now delivers an average of 17 times faster processing than the initial release without sacrificing accuracy from the previous 135sec median processing time down to 8sec. Users can now process hundreds to thousands of pages per minute.
• Accuracy: ADE parses structured forms and dynamic layouts for high-accuracy classification and extraction. It now delivers higher accuracy when extracting data from complex tables and multi-column layouts ensuring reliable results even with challenging formats.
• Trust: Visual Grounding links outputs to exact document locations (page, region, coordinates) for traceability, validation, and compliance. ADE now offers zero-data retention for enterprise customers.
• Economical: Whether for research or large-scale production, ADE streamlines complex document processing, reducing manual review and saving costs

A leading healthcare platform provider, Eolas Medical, is processing over 100,000 clinical guidelines in the form of PDFs and complex documents with ADE, streamlining the creation of structured summaries with the view to supporting over 1.2million queries per month from healthcare professionals on their platform. Their QA chatbot, powered by ADE, provides answers with direct references to the original documents, improving information traceability and reliability.

"ADE has significantly outperformed other document extractors we've used," said Declan Kelly, the CEO of Eolas Medical. "It has helped us build an Agentic RAG answer engine, based on unique healthcare institutional content, to offer instant, validated support to medical professionals at the point of care."

In financial services, ADE is being used to automate document onboarding for use cases like Know Your Customer (KYC), mortgage and loan processing, and client due diligence. Visual grounding enables full auditability by linking extracted data directly to its source location in the document

Dan Maloney, CEO at LandingAI, states, "We've witnessed firsthand the challenges businesses face when extracting meaningful data from complex documents. Agentic Document Extraction solves these challenges by providing a powerful, accurate, and efficient solution that unlocks valuable insights from documents. It's helping customers unlock millions of data points previously buried in forms and PDFs—at scale and with confidence."

ADE is now available through self-serve and enterprise plans. For enterprise customers, tailored pricing and customized request limits are available as well as zero-data retention is supported for privacy and compliance, with additional compliance features coming soon.

To access ADE API and try out on your documents, visit Agentic Document Extraction here.

About LandingAI:
LandingAI™ delivers cutting-edge agentic vision technologies that empower customers to unlock the value of visual data. With LandingAI's solutions, companies realize the value of AI and move AI projects from proof-of-concept to production.

LandingAI's flagship product, LandingLens™, enables users to build, iterate, and deploy vision AI solutions quickly and easily. LandingAI is a pioneer in agentic vision AI technologies, including Agentic Document Extraction and Agentic Object Detection, which enhance the ability to process and understand visual data at scale, making sophisticated Visual AI tools more accessible and efficient.

Founded by Andrew Ng, co-founder of Coursera, founding lead of Google Brain, and former chief scientist at Baidu, LandingAI is uniquely positioned to lead the development of Visual AI that benefits all. For more information, visit Landing.ai.

SOURCE LandingAI