Which document formats does DocSlice support?

DocSlice currently supports PDF, DOCX, RTF, TXT, CSV and images (PNG, JPG, WEBP, TIFF, BMP) with automatic OCR. The old binary .doc format is not supported: save it as .docx or PDF. XLSX and PPTX are planned.

Where is my data processed and how long is it kept?

DocSlice runs on EU-only infrastructure (Frankfurt, Germany). Uploaded files are deleted immediately after processing; async job results are kept for 24 hours exclusively for retrieval and then automatically purged. Documents are never used for training. See the privacy policy at https://docslice.app/privacy.html for details.

What is a custom JSON schema in DocSlice?

Users can define a JSON schema specifying exactly which fields to extract from a document — such as invoice number, date, total, and line items. The AI layer finds and returns only those fields.

Does DocSlice have a free plan?

Yes. DocSlice is free during the private beta. The indicative launch plan gives 500 welcome pages plus 50 pages every month that reset, with all output formats and the REST API; final prices are confirmed after the beta.

How does the DocSlice API work?

DocSlice provides a REST API. Developers send a document via POST request with their API key and desired output format or custom JSON schema. The API returns the structured data in the response, or later through an asynchronous job and a webhook.

Waitlist open · early access

Turn any document into structured data.

PDF, DOCX, TXT and more — DocSlice extracts, understands and delivers your content in the format you need. AI-powered. Privacy-first. EU infrastructure.

🇪🇺 EU Hosting No Model Training AI Layer REST API API Key in 30s

// join_waitlist.sh

Email address How will you use DocSlice?

No spam · No credit card · Free credits at launch

✓

You're on the list!

We'll reach out as soon as DocSlice is ready. Priority access + free credits guaranteed.

Developers and agencies already waiting

POST api.docslice.app/v1/convert

INPUT · invoice.pdf

OUTPUT · JSON

{
  "invoice_no": "INV-2847",
  "date": "2026-05-21",
  "total": 1240.00,
  "vat": 22,
  "items": [
    {"qty":2,
     "sku":"API-PRO"}
  ]
}

How it works

Three steps to structured data.

01 ——

📄

Upload your document

PDF, DOCX, RTF, TXT, CSV, images — drag and drop or send via API. Processed in Europe, deleted immediately after conversion.

02 ——

🎛️

Define your output

Pick a format (JSON, Markdown, HTML, TXT, DOCX) or define a custom schema. Tell DocSlice exactly which fields to extract.

03 ——

⚡

Receive clean data

The AI layer cleans, normalises and structures output. Ready for your database, CMS, n8n or Zapier workflow.

Why DocSlice

Built differently, by design.

🔒

Privacy-first by design

EU-only infrastructure. Uploaded files deleted immediately after parsing. Async results auto-purged after 24h. No training, no exceptions.

🤖

AI-powered extraction

Not just text parsing — an AI layer understands context, removes noise and delivers output based on modern OCR & Document AI (Wikipedia).

🎛️

Custom output schemas

You define the fields, DocSlice extracts and validates against the JSON Schema standard. Invoice data, contract clauses, product specs.

⚡

Developer-first API

Clean REST API following OpenAPI specifications, excellent docs, API key in 30 seconds. Works with n8n, Make and Zapier out of the box.

Pricing preview

Simple, transparent pricing.

Free

€0

No credit card

500 welcome pages (no expiry)
+ 50 pages every month, reset monthly
All output formats
REST API · EU hosting

Common questions.

About us

Built with care by the DocSlice Team.

DocSlice was born out of frustration with clunky OCR tools, brittle regex parsers, and black-box cloud APIs that compromise on data privacy. We are a European team of software engineers dedicated to providing developers, agencies, and modern enterprises with a fast, reliable, and privacy-first Document Intelligence API.

We keep as little data as possible: uploaded files are processed in Frankfurt (EU) and deleted immediately after parsing. For asynchronous jobs, extracted results are retained for up to 24 hours exclusively to allow retrieval before automated permanent purging. We never train on your data. We build on open standards like JSON (Wikipedia) and W3C standards.

Author & Team: DocSlice Engineering Team · Frankfurt & Remote EU · Meet the team & mission →

Contact: hello@docslice.app

Ready to slice through your documents?

Join the waitlist. Free credits at launch. No commitment.

Get early access →