Autype

Autype Lens

Turn incoming documents into usable data.

Read the files your business already receives, extract exactly what the next system needs, and keep the workflow moving without manual sorting or retyping.

Try Autype LensExplore the Lens API

One API for OCR, field extraction, classification, and file naming.

Solis Components BV

Supplier invoice

PDF

Invoice

Professional services · August 2026

Invoice no.

SC-2026-0718

Supplier

Solis Components BV

Due date

17 Aug 2026

DescriptionAmount
Implementation services€2,950.00
Support package€950.00
Travel expenses€348.30
Total due€4,248.30
Four focused operationsRecover content, classify, extract fields, or generate filenames.
Results your system can useReceive Markdown, typed JSON, categories, and predictable names.
Your existing filesProcess PDF, DOCX, ODT, and Markdown through one job model.

Choose the result the workflow needs

One file can answer four different questions.

Recover its content, decide where it belongs, extract specific values, or give it a useful name. Lens performs the focused operation you choose.

Extract fields

Return the exact data your system expects.

Define the fields and types once. Lens returns matching JSON and makes missing values explicit.

supplier-invoice.pdf
Structured resultfieldsMissing: []
{
  "invoiceNumber": "SC-2026-0718",
  "supplier": "Solis Components BV",
  "totalDue": 4248.30
}

Extraction that follows your schema

Ask for the values you need, not every word on the page.

Define field names, types, and optional descriptions. Lens returns matching values with the same keys and calls out anything it could not find.

Your schema

Up to 30 fields
invoiceNumberstring
totalAmountnumber
invoiceDatedate
purchaseOrderstring

Typed result

Complete

invoiceNumber

SC-2026-0718

totalAmount

4248.30

invoiceDate

2026-08-17

purchaseOrder

Not provided in this document

Optional

Automate document intake

Stop opening every file before the workflow can continue.

Lens can identify what arrived, extract the values the destination needs, and generate a consistent filename. Your workflow can route the result without a manual sorting queue.

Incoming documentsLens intake

File 01

Supplier invoice

invoice-SC-0718.pdf
Accounts payable

File 02

Service agreement

contract-Meridian-2026.pdf
Legal workspace

File 03

Research report

report-energy-outlook.pdf
Content migration

From fixed pages to reusable structure

Recover content you can edit, migrate, or render again.

For PDF input, Lens can return Autype Extended Markdown or full document JSON with sections, headings, tables, embedded content images, and document settings.

Keep the recovered source readable

Import the result into Autype or process it in code

Reuse document structure instead of starting over

report.pdf

Regional growth

Recovered document
Text and headings
Tables and figures
Page settings

Clean Markdown

Lightweight content for search, migration, summaries, and downstream text processing.

PDF, DOCX, ODT, Markdown

Extended Markdown

Readable source with Autype document settings and defaults that can be edited or rendered again.

PDF input

Document JSON

Sections, elements, and document structure for complete programmatic control.

PDF input

A focused document API

Add document understanding without building the pipeline yourself.

Upload a file, start the Lens operation, then read structured data from the completed job. The same authentication, status, and webhook model is used across Autype developer tools.

Read the Lens APIExplore the Document Engine
Extract typed fieldsPOST
/api/v1/dev/tools/lens/extract

{
  "fileId": "file_7f42",
  "fields": {
    "invoiceNumber": { "type": "string" },
    "totalAmount": { "type": "number" },
    "invoiceDate": { "type": "date" }
  },
  "pages": ["1-2"]
}

Typed JSON

Values follow your schema

Async result

Poll or receive a webhook

Same job model

Consistent with Autype tools

Autype Lens FAQ

Know exactly where Lens fits.

The practical product boundaries and workflow details before you connect it.

Is Lens just OCR?

No. OCR is one operation. Lens can also classify a document into categories you provide, generate a filename from its contents, and extract typed values against your field schema.

Which files can Lens process?

All four Lens operations accept PDF, DOCX, ODT, and Markdown. Standard Markdown recovery works across those formats. Autype Extended Markdown and full document JSON recovery are available for PDF input.

Can I process only selected pages?

Yes. Structured field extraction supports page selections for PDFs. Standard Markdown recovery can also focus on selected PDF pages, so you do not need to process an entire long document.

What happens when a requested field is missing?

Lens returns the field as null and lists its name in fieldsMissing. The next workflow step can handle missing information explicitly instead of receiving an invented value.

Can Lens route a document automatically?

Lens returns the category, extracted values, or generated filename. Your API workflow, n8n flow, Make scenario, or application can use that result to choose the next destination or action.

Make every incoming document useful.

Extract the content and data you need, then continue the workflow through Autype’s Document Engine and developer tools.

Start trialExplore the API

Use Lens through the same Autype account, API key, and job model.