Amazon Textract
Machine learning service that extracts text, forms, tables, and data from documents.
- Tool type
- Machine learning document extraction service
- Made for
- Document processing teams and developers
- Runs on
- AWS APIs and the AWS Management Console
- Applications call Textract APIs; the AWS Console is used to explore the service and train Custom Queries adapters.
What it does
Amazon Textract takes in scanned documents such as PDFs, images, forms, tables, invoices, receipts, identity documents, and mortgage packages. It returns extracted printed text and handwriting, along with structured output such as key-value pairs from forms, table data organized in rows and columns, answers to user-defined queries, detected signatures, and layout elements. The service is positioned as going beyond basic OCR by identifying and extracting specific data without manual configuration.
People typically use Textract by calling one of its APIs: Detect Document Text for OCR-style text extraction, Analyze Document for forms, tables, queries, and signatures (in any combination), Analyze Expense for invoices and receipts, Analyze ID for passports and driver's licenses, and Analyze Lending for mortgage application documents. Custom Queries lets a team train an adapter in the AWS Console for business-specific documents. The distinct value is the pretrained, ML-based extraction for common document types, with customization options when those are insufficient.
Textract suits teams in financial services, healthcare, and the public sector who process large volumes of forms, and developers who can call AWS APIs. Access requires an AWS account; the Console is used for setup and custom adapters. A free tier lasts three months with page allowances per API, and Custom Queries has no free tier. Pricing is per page and varies by feature, volume, and region, so high-volume use requires estimating costs with the AWS Pricing Calculator or a custom quote.
What it can do
- Extracts printed text and handwriting from scanned documents and images.
- Extracts key-value pairs from forms.
- Extracts tabular data organized in rows and columns.
- Answers user-defined natural-language queries about document content.
- Detects handwritten signatures, electronic signatures, and initials.
- Processes identity documents such as U.S. passports and driver's licenses to extract fields like date of birth and expiry date.
- Classifies, splits, and extracts data from mortgage application document packages through Analyze Lending.
Example workflows
- Mortgage application processing
- Input: a batch of scanned mortgage application pages. Steps: send pages to Analyze Lending, which classifies and splits document types and extracts results with a summary that notes whether a signature was detected. Output: classified document sections, extracted fields, and signature detection per page.
- Invoice data capture
- Input: scanned invoices or receipts. Steps: send each document to the Analyze Expense API, which returns OCR text along with extracted expense fields. Output: structured invoice data that a downstream system can store or review.
Before you choose
- Custom Queries has no free tier, and the free allowances for other APIs are limited to three months for new AWS customers.
- Pricing is per page and depends on the API, feature combination, monthly volume, and region; estimate costs with the AWS Pricing Calculator or request a custom quote for high-volume use.
- Requires an AWS account and familiarity with AWS APIs or the Console to build a working pipeline.
Listed under
Get started
New AWS customers can sign up for a free account to access the AWS Free Tier, then start building in the Console.
- Create or sign in to an AWS account via the AWS sign-up page.
- Open the Amazon Textract console to explore the service and choose the API that matches your document type.
- If your documents need business-specific extraction, train a Custom Queries adapter in the AWS Console and use its adapter identifier in your Analyze Document API requests.
Plans and prices
Free tier is a three-month allowance for new AWS customers with per-API page limits; it is not a permanent free tier. Custom Queries has no free tier. Costs scale with page volume and feature selection, with tiered rates above certain monthly thresholds. Use the AWS Pricing Calculator for estimates.
- Rates vary by API and feature combination
- Tiered pricing by monthly page volume
Source: aws.amazon.com/textract/pricing, checked 28 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.
- Intelligently Extract Text & Data with OCR - Amazon Textract - Amazon Web Services ↗aws.amazon.com
- Textract Pricing Page ↗aws.amazon.com
Ratings and reviews
No ratings for Amazon Textract yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- aws.amazon.com: product, pricing and policy pages
- YC directory mirror
- Batch and company name
- Web search
- Confirms the official site
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.