# PDF to Markdown Converter

> Turn reports, papers, manuals, and contracts into clean Markdown built for LLMs and RAG — not a flattened text dump.

## Why copy-paste fails

- Reading order breaks — Multi-column PDFs interleave columns, headers, and footnotes, so the text a model reads no longer matches the page.
- Tables flatten — Copy-paste collapses tables into loose runs of numbers and drops the heading hierarchy.
- Scans carry no text — Image-only pages have no text layer until they pass through OCR (an enhancement on the roadmap).

## Format notes

- Reading order reconstruction — Multi-column layouts are reordered left-to-right, top-to-bottom so the Markdown reads as a logical document rather than interleaved column fragments.
- Tables and headings — Heading levels and tabular data are preserved where the PDF exposes structural metadata; flat-text PDFs produce flat-text output.
- Scanned and image-only PDFs — PDFs without a text layer (scanned pages, image inserts) require OCR to extract content; higher-accuracy OCR is on the enhancement roadmap.
- Password-protected files — Encrypted PDFs cannot be parsed; remove the password before uploading.

## Example output

Source: quarterly-report.pdf

```markdown
# Quarterly Report

## Revenue

| Quarter | Revenue |
| --- | --- |
| Q1 | $1.2M |
| Q2 | $1.6M |
```

## FAQ

### Is it free?

Yes. Converting a PDF in the browser is free with no sign-up. Downloading the .md file and saving history require a free account.

### What is the maximum file size?

The free converter accepts PDF files up to 10 MB. Larger files are supported on paid plans.

### Does it handle scanned PDFs?

Image-based PDFs are supported through the parsing pipeline; higher-accuracy OCR is on the roadmap for better fidelity on scanned pages.

### Are multi-column layouts handled?

Yes. The parser reconstructs reading order so multi-column text flows as a single readable document rather than interleaved column fragments.

### Will tables be preserved?

Tables in PDFs that carry structural metadata are converted to Markdown tables. Flattened-image tables require OCR to recover.

### Can I convert password-protected PDFs?

No. Encrypted PDFs cannot be parsed. Remove the password before uploading.

### Is the output good for RAG?

Yes. The Markdown preserves headings, lists, and tables where possible so chunking and embedding stay predictable.

### Is there a developer API?

Yes. You can convert PDFs programmatically via the REST API. See /developers for documentation and API keys.