Skip to content

Google Document AI OCR Review

Google Document AI's Enterprise Document OCR is a Google Cloud API that extracts printed and handwritten text from PDFs and images, priced per page with the first 1,000 pages free.

Free plan · from $0.60 · API

Stork's verdict on Google Document AI OCR

Google's Enterprise Document OCR is a cheap, accurate base layer for high-volume text extraction, but structured data costs extra through Form or Layout Parser.

Google Document AI OCR reviewed by Stork AI · stork.ai/en/google-document-ai-ocr

720p. 480p is half that. Minimum 5 cents a video.
Ten minutes of talking video costs $2.
We run open models on our own GPUs. When our cost per minute falls, so does the price.
Make a photo talkDevelopers: there is an API

What Google Document AI OCR does

Enterprise Document OCR is one of the processors in Google Cloud's Document AI. You send a PDF or image (PDF, GIF, TIFF, JPEG, PNG, BMP or WebP; DOCX is in private preview) and get back a Document object in JSON with the text, pages, blocks, lines, tokens and their positions. It reads printed and handwritten text, can merge embedded text from digital PDFs with OCR results (native PDF parsing), and scores each page's image quality from 0 to 1, listing likely defects such as blur, glare or cut-off text when the score is below 0.5. Google says it handles more than 200 languages; handwriting is marked unsupported for a dozen of them, including Hebrew, Thai, Tamil and Telugu.

Setup is the usual Google Cloud flow: create a project, enable the Document AI API, create an Enterprise Document OCR processor in a region (US, EU and several others) and call it from the REST API or client libraries. Online (synchronous) requests take up to 15 pages and 40 MB (30 pages in imageless mode); batch requests take up to 500 pages per document, 1 GB per file and 5,000 files per request. Stable model versions are pretrained-ocr-v2.0-2023-06-02 and pretrained-ocr-v2.1-2024-08-07, with v2.1.1 as a release candidate outside the US and EU.

Google Document AI OCR pricing

Pay as you go, per 1,000 pages, billed per monthly cycle.

PlanPriceIncluded
Enterprise Document OCR, first 1,000 pages$0Free tier
Enterprise Document OCR, 1,000 to 5,000,000 pages$1.50 per 1,000 pagesText, handwriting, layout, image quality
Enterprise Document OCR, over 5,000,000 pages$0.60 per 1,000 pagesSame
OCR add-ons$6.00 per 1,000 pagesMath OCR, checkbox extraction, font-style detection (v2 processor only)

Example: 100,000 pages of plain OCR in a month is 1,000 free plus 99,000 at $1.50 per 1,000 = 99 x $1.50 = $148.50. With add-ons switched on for all of them, add 100 x $6.00 = $600 (the pricing page does not say whether the free tier covers add-ons).

Other Document AI processors are priced separately if you need structure rather than raw text:

ProcessorPrice per 1,000 pages
Layout Parser (includes chunking)$10.00
Form Parser$30.00 (up to 1M pages), $20.00 above
Custom extractor$30.00 (up to 1M pages), $20.00 above
Custom classifier / splitter$5.00 (up to 1M pages), $3.00 above

Gemini Enterprise Flexible Savings Plans cut these rates (OCR is $1.35 per 1,000 pages on a 1-year plan and $1.20 on a 3-year plan). Failed requests (4xx or 5xx) are not billed. A page means one PDF page or one image; for DOCX, up to 3,000 characters counts as a page. New Google Cloud customers get $300 in free credits under the Google Cloud Free Program.

Prices checked on the vendor's pricing page on November 10, 2026. See Document AI pricing.

Key features

What Google's Document AI documentation lists for Enterprise Document OCR:

  • Printed and handwritten text in more than 200 languages, with BCP-47 language hints
  • Native PDF parsing that merges embedded digital text with OCR
  • Image quality scores (0 to 1) with eight defect types such as blur, glare and text cutoff
  • Math OCR add-on: formulas returned as LaTeX with bounding boxes
  • Checkbox extraction add-on: filled and unfilled checkboxes and radio buttons
  • Font-style add-on: word-level font type, weight, color, bold and italic
  • Symbol-level output, page range selection and character-in-the-box detection
  • Batch jobs of up to 5,000 files and 500 pages per document

Who Google Document AI OCR is best for

It suits developers and data teams on Google Cloud who need high-volume text extraction from scans, faxes, forms and photos, and want per-page pricing that drops sharply above 5 million pages a month. The image-quality score is useful for routing unreadable scans back to a human, and the math and checkbox add-ons cover cases plain OCR misses. It is an API, not an app: there is no upload-and-download interface for occasional users.

If you need tables and key-value pairs rather than raw text, compare Google's own Form Parser or Layout Parser with Amazon Textract and Azure AI Document Intelligence, especially if you already run on AWS or Azure. For LLM-ready Markdown from complex PDFs, look at LlamaParse. For free, self-hosted OCR with no per-page bill, Tesseract is the open-source default.

What users say

Published comparisons put Google's OCR among the cloud leaders but rarely test Enterprise Document OCR on its own. A 2026 comparison from ImageToTable.ai, which sells a competing extraction tool and ran no tests of its own, cites outside benchmarks of about 95% accuracy on printed text and recommends choosing whichever cloud you already use. It also reports Google behind Azure and Textract on invoice line items and tables, citing a 2025 invoice benchmark. The common advice is to run a bake-off on your own documents before committing.

Pros

  • Low per-page price that falls to $0.60 per 1,000 above 5 million pages
  • First 1,000 pages free, plus $300 credits for new Google Cloud accounts
  • Handles handwriting and more than 200 languages
  • Image quality scores help catch unreadable scans
  • Math, checkbox and font-style add-ons on the v2 processor

Cons

  • API only: needs a Google Cloud project and code to use
  • Online requests capped at 15 pages; longer files need batch jobs
  • Tables and key-value fields require pricier processors ($10 to $30 per 1,000)
  • Handwriting unsupported in a dozen languages including Hebrew and Thai

Frequently asked questions

How much does Google Document AI OCR cost?

Enterprise Document OCR costs $1.50 per 1,000 pages from 1,000 to 5 million pages a month and $0.60 per 1,000 pages above that; the first 1,000 pages are free. OCR add-ons cost $6.00 per 1,000 pages. Prices checked November 10, 2026.

Is Google Document AI OCR free?

Partly. The pricing table lists the first 1,000 pages as free, and new Google Cloud customers get $300 in free credits. After that you pay per page; there is no unlimited free plan.

What file types does Enterprise Document OCR support?

PDF, GIF, TIFF, JPEG, PNG, BMP and WebP. DOCX support is in private preview, up to 15 pages in synchronous and 30 in asynchronous requests.

Can Google Document AI read handwriting?

Yes, for most supported languages. Google's language table marks handwriting as unsupported for Armenian, Gujarati, Hebrew, Kannada, Khmer, Lao, Malayalam, Punjabi, Tamil, Telugu, Thai and Yiddish.

What are the page limits per request?

Online requests take up to 15 pages and 40 MB (30 pages in imageless mode). Batch requests take up to 500 pages per document, files up to 1 GB, and up to 5,000 files per request.

What is the difference between Document AI OCR and Form Parser?

Enterprise Document OCR returns text and layout at $1.50 per 1,000 pages. Form Parser also extracts key-value pairs, tables and checkboxes as structured fields, at $30 per 1,000 pages up to 1 million pages. Layout Parser, at $10 per 1,000 pages, returns document structure and chunks for retrieval.

How does it compare with the Cloud Vision API?

Both come from Google Cloud. Document AI's Enterprise Document OCR is the document-focused processor, with image quality scores, native PDF parsing and add-ons such as math OCR and checkbox extraction, plus batch processing of multi-page documents.

Compare Alternatives

See every Google Document AI OCR alternative, compared

Google Document AI OCR

Free plan · from $0.60 · API

Enjoying this? Get one like it in your inbox each morning.

one email a day · unsubscribe in two clicks · no third-party tracking