Stay organized with collections
Save and categorize content based on your preferences.
Document AI pricing
This document explains Document AI pricing details.
Document AI is a solution and intended to be used with
other Google Cloud products. You might need to review the pricing
for Cloud Vision,
Cloud Natural Language API, or
Vertex AI.
You are not billed for failed requests (4xx or 5xx response codes).
If you pay in a currency other than USD, the prices listed in your currency on
Cloud Platform SKUs
apply.
To parse a document with between 1 to 10 pages costs $0.10.
To parse a document with between 11 to 20 pages costs $0.20.
To parse a document with between 91 to 100 pages costs $1.
Note that documents with more than 10 pages are not supported by synchronous
requests. You can process multiple documents, each up to 200 pages, by sending
a batch request. For more information, see
Limits for specialized processors.
If you pay in a currency other than USD, the prices listed in your currency on
Cloud Platform SKUs
apply.
If the input document has multiple eligible forms, only the first is
processed and charged the price listed. If the input document has no
eligible form, processing doesn't occur and no charge is made.
Billing questions
Use the following resources to get help with billing questions:
To change or disable billing on a project, go to the Billing page in the
Google Cloud console. For more information, see
Modify a Project's Billing Settings
in the Google Cloud console Help.
[1] The cost to call the custom extractor is the same whether you use a generative AI-powered processor version or custom model.
[2] Only available for Enterprise Document OCR processor (v2).
[3] This processor is only available to limited access
customers.
To request API access, fill out and submit the
Document AI limited
access customer request form.
The form requests information about you, your company, and your use case.
Note that a Google Cloud Project ID is required for access.
To create a new Google Cloud project, or identify your existing project's
Project ID see the following
instructions.
After you submit the form, the Document AI team will
review your request to ensure you meet the criteria for access.
If approved, you will receive an email with instructions on how to access
and use this feature.
[4] You are not billed if a document is classified as "other," by Procurement
Document Splitter & Classifier instead of a specific document type.
[5] The size of a page depends on the file format.
Images (JPEG/JPG, PNG, BMP, HEIF): Each image = 1 page
PDF: Each page in the PDF = 1 page
TIFF: Each image in the TIFF = 1 page
Word (DOCX): Up to 3,000 characters = 1 page
Excel (XLSX): Each tab = 1 page
Powerpoint (PPTX): Each slide = 1 page
HTML: Up to 3,000 characters = 1 page
Parsed Documents: Up to 3,000 characters = 1 page
Request a custom quote
With Google Cloud's pay-as-you-go pricing, you only pay for the services you
use. Connect with our sales team to get a custom quote for your organization.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],[],[[["\u003cp\u003eDocument AI pricing is based on usage, with costs varying by processor type and monthly page volume, and is designed to be used with other Google Cloud products.\u003c/p\u003e\n"],["\u003cp\u003eThe cost for digitizing text with the Enterprise Document OCR Processor ranges from $0.60 to $1.50 per 1,000 pages depending on volume, and the OCR add-ons are priced at $6 per 1,000 pages.\u003c/p\u003e\n"],["\u003cp\u003eExtracting data from documents through custom extractors, Form Parser, or Layout Parser costs between $10 and $30 per 1,000 pages, with volume-based discounts, and re-chunking is priced at $0.02 per 1,000 pages.\u003c/p\u003e\n"],["\u003cp\u003eFine-tuning generative AI models in the US region costs $46.39 per hour, which includes the cost of the underlying Vertex AI machines, while there is no cost for training custom models.\u003c/p\u003e\n"],["\u003cp\u003ePretrained processors for tasks like invoice parsing and document classification are priced per document or per 10 pages, ranging from $0.05 to $0.30, and custom processor hosting is $0.05 per hour per deployed version.\u003c/p\u003e\n"]]],[],null,["# Pricing\n\nDocument AI pricing\n===================\n\nThis document explains Document AI pricing details.\n\nDocument AI is a solution and intended to be used with\nother Google Cloud products. You might need to review the pricing\nfor [Cloud Vision](/vision/pricing),\n[Cloud Natural Language API](/natural-language/pricing), or\n[Vertex AI](/vertex-ai/pricing#automl_models).\nYou are not billed for failed requests (`4xx` or `5xx` response codes).\n\n\nIf you pay in a currency other than USD, the prices listed in your currency on\n[Cloud Platform SKUs](https://cloud.google.com/skus/)\napply.\n\nDigitize text\n-------------\n\nExtract structures and entities from documents\n----------------------------------------------\n\nBreak documents into chunks\n---------------------------\n\nIf you pay in a currency other than USD, the prices listed in your currency on [Cloud Platform SKUs](https://cloud.google.com/skus/) apply.\n\n### Pricing examples\n\n#### Example 1\n\nYou sent 100 pages to Form Parser in your monthly billing cycle. Your\nmonthly bill is:\n\n- ($30 / 1,000 pages) \\* (100 pages) = $3 for prediction services\n\n#### Example 2\n\nYou sent 6 million pages to Form Parser in your monthly billing cycle, your\nmonthly bill is:\n\n- ($20 / 1,000 pages) \\* (5,000,000 pages) = $100,000 for prediction services\n- ($30 / 1,000 pages) \\* (1,000,000 pages) = $30,000 for prediction services\n- Total: $130,000\n\nClassify documents\n------------------\n\nCustom processor hosting\n------------------------\n\nWith custom processors, you pay for hosting and prediction.\n\n### Hosting charges\n\nPricing example:\n\nOne processor version deployed for a year costs $438 ($.05 \\* 24 \\* 365).\n\nPretrained processor charges\n----------------------------\n\n### Pricing examples\n\nHere are a few pricing examples:\n\n- To parse a document with between 1 to 10 pages costs $0.10.\n- To parse a document with between 11 to 20 pages costs $0.20.\n- To parse a document with between 91 to 100 pages costs $1.\n\nNote that documents with more than 10 pages are not supported by synchronous\nrequests. You can process multiple documents, each up to 200 pages, by sending\na batch request. For more information, see\n[Limits for specialized processors](/document-ai/quotas#specialized_processors).\n\nIf you pay in a currency other than USD, the prices listed in your currency on [Cloud Platform SKUs](https://cloud.google.com/skus/) apply.\n\nIf the input document has multiple eligible forms, only the first is\nprocessed and charged the price listed. If the input document has no\neligible form, processing doesn't occur and no charge is made.\n\nBilling questions\n-----------------\n\nUse the following resources to get help with billing questions:\n\n- To learn more about billing, read the [Cloud Billing documentation](/billing/docs).\n- Resolve billing concerns by using the [billing concerns troubleshooter](https://support.google.com/cloud/troubleshooter/7279311).\n- Request help with billing questions by using the [billing support form](https://support.google.com/cloud/contact/cloud_platform_billing).\n- To change or disable billing on a project, go to the **Billing** page in the Google Cloud console. For more information, see [Modify a Project's Billing Settings](https://support.google.com/cloud/answer/6293499) in the Google Cloud console Help.\n\n| **Caution:** When you disable billing for a project, you also disable all product resources in that project.\n\nWhat's next\n-----------\n\n- Read the [Document AI documentation](/document-ai/docs/overview).\n- Get started with [Document AI](/document-ai/docs/how-to).\n- Try the [Pricing calculator](/products/calculator).\n- Learn about [Document AI solutions and use cases](/document-ai#section-6).\n\nNotes\n-----\n\n\n\\[1\\] The cost to call the custom extractor is the same whether you use a generative AI-powered processor version or custom model.\n\n\n\\[2\\] Only available for Enterprise Document OCR processor (v2).\n\n\n\\[3\\] This processor is only available to limited access\ncustomers.\n\nTo request API access, fill out and submit the\n[Document AI limited\naccess customer request form](https://docs.google.com/forms/d/e/1FAIpQLSc_6s8jsHLZWWE0aSX0bdmk24XDoPiE_oq5enDApLcp1VKJ-Q/viewform?gxids=7826).\nThe form requests information about you, your company, and your use case.\nNote that a Google Cloud Project ID is required for access.\nTo create a new Google Cloud project, or identify your existing project's\nProject ID see the following\n[instructions](https://cloud.google.com/resource-manager/docs/creating-managing-projects).\n\nAfter you submit the form, the Document AI team will\nreview your request to ensure you meet the criteria for access.\nIf approved, you will receive an email with instructions on how to access\nand use this feature.\n\n\u003cbr /\u003e\n\n\n\\[4\\] You are not billed if a document is classified as \"other,\" by Procurement\nDocument Splitter \\& Classifier instead of a specific document type.\n\n\n\\[5\\] The size of a page depends on the file format.\n\n- Images (JPEG/JPG, PNG, BMP, HEIF): Each image = 1 page\n- PDF: Each page in the PDF = 1 page\n- TIFF: Each image in the TIFF = 1 page\n- Word (DOCX): Up to 3,000 characters = 1 page\n- Excel (XLSX): Each tab = 1 page\n- Powerpoint (PPTX): Each slide = 1 page\n- HTML: Up to 3,000 characters = 1 page\n- Parsed Documents: Up to 3,000 characters = 1 page\n\n#### Request a custom quote\n\nWith Google Cloud's pay-as-you-go pricing, you only pay for the services you use. Connect with our sales team to get a custom quote for your organization.\n[Contact sales](/contact?direct=true)"]]
