Docen
← BLOG·RELEASE

Introducing Docen Parse

Our core parsing model turns any document into clean, ordered Markdown, HTML, and JSON.

THE DOCEN TEAM·Sep 24, 2024·5 MIN READ

Today we're introducing Docen Parse: the core model behind everything we do. It turns a document — scan, PDF, or photo — into clean, ordered Markdown, HTML, and JSON, with the layout intact.

Why parsing first

Almost every document workflow starts with the same step: get the text out, in the right order, with the structure preserved. Do that badly and everything downstream inherits the mess. Do it well and extraction, search, and analysis all get easier.

Parse reads a page the way a person does — following columns, respecting headings, keeping tables as tables — and returns something you can build on.

  • Reading order across columns and pages
  • Tables preserved as structured cells
  • Math, lists, and code kept faithful
  • One call from file to clean text

What's next

Parse is the foundation. Recognition, layout, extraction, and evaluation all build on top of it, and we'll be shipping those in the months ahead.

Want to see it on your own documents? Open the playground or reach out — we're happy to run a sample with you.

parsingMarkdownstructure
[]TRY DOCEN

Run your hardest documentthrough Docen.

See the structured output for yourself, or reach the team at support@docen.co.