Data Extraction API

Flexible plans for document parsing at scale

Parse documents into spatial JSON or Markdown, or extract specific fields to a schema — with confidence scores and source coordinates. Start free and choose the mode that fits your workflow.

Loading...

How credits work

Every operation is billed per page. Some operations are priced by processing mode; others use a single flat rate.

Parse

Priced by mode

Convert documents into text or structured output.

Mode Credits/page
Text
1
Structure
1.5
Understand
9
Agentic
18

Extract

Priced by mode

Pull specific fields to a schema. Rates include the Parse component.

Mode Credits/page
Text
7
Structure
7.5
Understand
15
Agentic
24

Classify

Flat rate

Sort documents into predefined categories.

Mode Credits/page
Default
1

Estimate monthly credits

Enter your expected document volume and request type.


Parse mode
Text
Request type
Parse
Pages per document
Documents per month
Estimated monthly credits 0
Recommended plan Free


Frequently asked questions

What is the Data Extraction API?

The Data Extraction API parses PDFs, scans, images, and Office files into structured output, including spatial JSON and Markdown, or extracts specific fields to a schema with confidence scores and source coordinates.

What is a credit?

Credits are the unit of consumption for the Data Extraction API. Parse requests cost 1 credit per page in text mode, 1.5 credits per page in structure mode, 9 credits per page in understand mode, and 18 credits per page in agentic mode. Extract requests add 6 credits per page to the selected Parse mode, for totals of 7, 7.5, 15, or 24 credits per page. Classify requests cost a flat 1 credit per page, regardless of Parse mode. Unused credits do not roll over to the next month.

Which mode should I use to estimate my credit usage?

Understand mode is the recommended default — it handles all document types with structured extraction and semantic enrichment. For Parse, estimate monthly usage as 1 credit per page in text mode, 1.5 credits per page in structure mode, 9 credits per page in understand mode, or 18 credits per page in agentic mode. For Extract, add 6 credits per page to the selected mode. For Classify, estimate a flat 1 credit per page.

What is the difference between monthly and annual billing?

Monthly billing charges month-to-month with no commitment. Annual billing is paid upfront for 12 months at a discounted rate.

What happens if I exceed my monthly credit limit?

With pay-as-you-go, processing doesn’t stop when your monthly credits run out. We pre-authorize charges in $25 increments and add credits as you use them, up to a spending cap you control (by default, half your monthly subscription value) — we never charge beyond it. Set the cap to $0 to turn pay-as-you-go off and stop at your monthly limit. You can also upgrade to a higher plan or contact sales for a custom quota.

Is there a free trial?

Yes. The Free plan gives you 5,000 credits/month with no credit card required — enough for 5,000 text Parse pages, about 3,333 structure Parse pages, or about 555 understand Parse pages. For Extract, the same 5,000 credits cover about 714 text + Extract pages, 666 structure + Extract pages, 333 understand + Extract pages, or 208 agentic + Extract pages. For Classify, the same 5,000 credits cover 5,000 pages.

Can I upgrade or downgrade my plan?

Yes. Upgrades take effect immediately. Downgrades take effect at the start of your next billing cycle.

Is the Data Extraction API secure and compliant?

Yes. The Data Extraction API is built with enterprise-grade security and processes documents in a secure cloud environment. It supports compliance requirements such as GDPR, HIPAA, and SOC 2 Type 2-audited controls.

What support is included with the Data Extraction API?

All plans include access to documentation and developer support. Additional support options may be available depending on your plan.

Ready to try it?

Create an account to get your DWS Processor API key and start making API calls.