What is DocSlice?

DocSlice is a REST API for document intelligence: you upload a document and it gives you back its content as structured data. There is nothing to install and no model to host; you make an HTTP request and read the answer.

What it does

  • Converts PDF, DOCX, RTF, TXT, CSV, LOG files and images (PNG, JPG, JPEG, WEBP, TIFF, BMP) into json, markdown or html.
  • Extracts fields you define. Describe what you want with a JSON Schema (for example invoice_number, date, total) and DocSlice returns an object that follows it.
  • Reads scans and photos. When a PDF has no usable text, or when you send an image, DocSlice applies OCR automatically and tells you so with ocr_applied: true.
  • Scales from one file to many. Send a single document, a batch of files or a ZIP archive, synchronously or as a background job with a webhook when it finishes.

Who it is for

Developers and teams who need document content inside their own software: accounting and expense tools reading invoices, back-office automations reading contracts, catalog importers reading product sheets, and anyone preparing text for search or for a language model.

What makes it different

  • Processed in Europe. Documents are processed in Frankfurt, Germany, and the AI and OCR provider is European.
  • Nothing is kept longer than needed. Uploaded files are deleted as soon as they have been processed. Results of background jobs are erased after 24 hours. Your documents are never used to train models.
  • One simple contract. One endpoint for conversion, plain JSON answers, and the same API key for everything.

What it is not

DocSlice is not a document store. It does not keep your files and it does not keep results for long: save what you need on your side. See How it works for the details of what happens to a document.