Learn how to build the next generation of AI applications. Join the Applied AI Summit on December 13
Jump to
Vertex AI's document processing solution

Document AI

Structure document data that you can store, analyze, search, and use to automate processes. Document AI extracts data from, classifies, and splits documents through a suite of pretrained models or through Workbench custom models. Finally, it uses Warehouse to search and store documents. 

  • Manage the entire unstructured document lifecycle in one unified solution

  • Reduce manual document processing, minimize setup costs, and accelerate deployment

  • Ensure a high level of accuracy with Google's AI and Human-in-the-Loop (HITL) reviews

  • Use your document data to gain new insights about your products and meet customer expectations

  • Use generative AI to easily extract data, search, and summarize documents  


Cost-effective and flexible

Improve operational efficiency by extracting structured data from unstructured documents and making that structured data available to your business apps and users.

Ensure your data is accurate and compliant

Automate and validate all your documents to streamline compliance workflows, reduce guesswork, and keep data accurate and compliant.

Use your data to meet customer expectations

Leverage insights to meet customer expectations and improve CSAT, advocacy, lifetime value, and spend.


Try Document AI in your environment

Upload a document (like an invoice) and see the structured data extracted. Don't have a document? Try our sample.

Key features

A unified platform to meet all your document processing needs

Process documents from a unified console

The Document AI platform is a unified console for document processing that lets you quickly access all models and tools. Use Document AI's pretrained models for document processing, including basic extractors like OCR and Form Parser, and specialized models for industry use cases like lending, contracts, procurement, and identity documents. Or use Document AI Workbench to uptrain these models on your business documents or create your own models and get better results for documents from your organization. With Document AI Warehouse, you can search, store, and manage documents, and even trigger workflows. Document AI lets you automate and validate documents to streamline workflows, reduce guesswork, and keep data accurate and compliant.

Leverage Google's state-of-the-art technologies

Document AI is built on decades of AI innovation at Google, bringing powerful and useful solutions to these challenges. Under the hood are Google’s industry-leading technologies: computer vision (including OCR), foundation models, and natural language processing (NLP) that create pretrained models for high-value, high-volume documents. The latest ML research and toolkits, which power Document Workbench and semantic search, are what makes Document Warehouse so much better than traditional document repositories. 

Generative AI for faster, simpler, improved results

New foundation model integration into Document AI Workbench helps you quickly improve custom processors through prompts. For example, you can add a new field to your data by prompting a foundation model to add this new field to your data instead of having to label and train a new model. You can also use the same approach to auto label new datasets. Easily generate summaries for your documents and customize them (long or short or others) based on your preferences. And, in Document AI Warehouse, get answers to natural language questions across a corpus of documents using generative AI, with fine-grained access controls.

Enrich data to make it more useful

Validate and enrich parsed information with Google knowledge graph technology to make the data even more useful, checking company names, addresses, phone numbers, and other details against entities on the internet.

Integrate human review into ML predictions

Human-in-the-Loop AI is a new DocAI feature that will help companies achieve higher document processing accuracy with the assurance of human review. Adding human review can increase accuracy and help businesses interpret predictions using purpose-built tools to enable those reviews.