Document AI Custom Extractor is now GA!  Try the demo below
Jump to

Document AI

Create document processors that help automate tedious tasks, improve data extraction, and gain deeper insights from unstructured or structured document information. Document AI helps developers create high-accuracy processors to extract, classify, and split documents.

  • Use generative AI to extract and structure data in minutes

  • Seamlessly connect to BigQuery, Vertex Search, and other Google Cloud products 

  • Enterprise-ready, along with Google Cloud's data security and privacy commitments

  • Built for developers; use the UI or API to easily create document processors


Faster time to value

Use generative AI to extract data or classify documents out of the box, with no training necessary to get started. Simply post a document to an enterprise-ready API endpoint to get structured data in return.

Higher accuracy

Document AI is powered by the latest foundation models, tuned for document tasks. Also, with powerful fine-tuning and auto-labeling features, the platform offers multiple paths to reach the required accuracy.

Better decision-making

Structure and digitize information from documents to drive deeper insights using generative AI to help businesses make better decisions.


Try Document AI in your environment

Extract data from your documents using generative AI.  For full product capabilities head to Document AI in the Google Cloud Console.

Key features

Use generative AI for document processing

Document AI Workbench

Document AI Workbench provides an easy way to build custom processors to classify, split, and extract structured data from documents. Workbench is powered by generative AI, which means it can be used out of the box to get accurate results across a wide array of documents. Furthermore, you can achieve higher accuracy by providing as few as 10 documents to fine-tune the large model—all with a simple click of a button or an API call. 

Try it now or learn more.

Enterprise OCR

With Enterprise Document OCR, users gain access to 25 years of optical character recognition (OCR) research at Google. OCR is powered by models trained on business documents and can detect text in PDFs and images of scanned documents in 200+ languages. The product can see the structure of a document to identify layout characteristics like blocks, paragraphs, lines, words, and symbols. Advanced features include best-in-class handwriting recognition (50 languages), recognizing math formulas, detecting font-style information, and extracting selection marks like checkboxes and radio buttons.

Try Document OCR now for accurate text and layout extraction.

Form Parser

Developers use Form Parser to capture fields and values from standard forms, to extract generic entities, including names, addresses, and prices, and to structure data contained in tables. This product works out of the box and does not require any training or customization and is useful across a broad range of document customization.

Explore document processing with Form Parser.


Try out pretrained models for commonly used document types including W2, paystub, bank statement, invoice, expense, US driver license, US passport, and identity proofing.

Explore pretrained options in the processor gallery.


Higher processing accuracy and faster development time

Document AI is helping customers improve fraud detection, automate customer support, and process clinical trial data.

What's new

See the latest updates about Document AI

Sign up for Google Cloud newsletters to receive product updates, event information, special offers, and more.



Google Cloud Basics
Document AI overview

Get an overview of the basics of Document AI, including extracting text from documents, classifying documents, and entity extraction.

Document AI introduction videos and labs

Get started learning about Document AI with our video series "The Future of Documents" and step-by-step codelabs.

Setting up the Document AI API

This guide provides all required setup steps to start using Document AI.

Use cases

Use cases

Use case
Extract data to drive automation and analytics

Use Document AI Workbench to automate data entry by extracting structured data from your documents. Typical applications include the mail room, shipping yards, mortgage processing divisions, procurement, and more. Use this data to make more efficient and effective business decisions.

Try out the Custom Extractor

Use case
Uncover insights buried in documents with BigQuery

You can now extract metadata from documents directly into a BigQuery objects table. Seamlessly join the parsed data with other BigQuery tables to combine structured and unstructured data, paving the way for comprehensive document analytics. 

Learn more about the BigQuery and Document AI integration here

Use case
Classify documents

Assigning categories or classes to documents as they flow into a business process makes them easier to manage, search, filter, or analyze. Custom Splitter and Classifier use machine learning to accurately predict and categorize single documents or multiple documents within a file. Use these products to improve efficiencies of the document processes.

Learn more and try out the Custom Splitter and Custom Classifier.

Use case
Make document processing applications smarter

SaaS customers and ISV partners can improve and expand their document processing solutions, quickly with generative AI. With a simple API prediction endpoint and document response format, customers can take document applications to the next level.

Use case
Digitize text for ML model training

Enterprise Document OCR enables users to create value from archival content that is otherwise unusable for training machine learning models. OCR helps extract text from scanned documents, plots, reports, and presentations prior to saving on a cloud storage or a data warehouse. Use these high-quality OCR outputs to accelerate your digital transformation initiatives such as training ML models specific to your business. 

Use case
Expand business capabilities with generative AI

Capture document information for new generative AI architectures and frameworks. Combining OCR and the Vertex AI PaLM API lets users unlock valuable data from documents to build document Q&A experiences, perform automated document comparison, or even generate new documents.   


Document AI pricing

Document AI offers transparent, cost effective pricing for all your document processing, model training, and storage needs. Visit our pricing page for more details.

If you pay in a currency other than USD, the prices listed in your currency on Google Cloud SKUs apply.


Document AI partners

Get help implementing Document AI from these trusted partners. View full partner directory.