PDF to HTML
Convert selected PDF pages into a standalone local HTML file. Semantic mode creates accessible headings, paragraphs, lists, and tables; layout mode positions text at reconstructed page coordinates. Both outputs include a restrictive content security policy and no PDF JavaScript.
- Processed locally in your browser
- No watermark
- No sign-up
- Works offline
Fidelity mode: Choose responsive semantic HTML for accessibility and editing, or positioned layout HTML for closer text geometry. Neither mode executes PDF JavaScript.
Choose a PDF or drop it here
One PDF · layout analysis and output generation stay on this device
Output compatibility
semantic or positionedPDF structure is reconstructed heuristically; this report states the selected fidelity trade-off.
Preserved
- searchable text
- heuristic structure
- page grouping
- safe standalone output
Approximated
- reading order
- fonts
- semantic roles or glyph placement
Omitted
- embedded images
- complex vectors
- forms
- executable PDF JavaScript
How it works
- 1Open a PDF and choose pages, reading order, repeated-margin cleanup, and optional local OCR.
- 2Select semantic reflow for accessible content or positioned layout for closer page geometry.
- 3Review which structure is preserved, approximated, or omitted.
- 4Download a standalone inert HTML document with no active PDF content.
Your privacy
This tool runs entirely inside your browser using JavaScript and Web Workers. Your document is never uploaded to any server — you can even disconnect from the internet and keep working once the page has loaded.
Frequently asked questions
▸Which mode should I choose?
Semantic mode is better for accessibility, responsive reading, and editing. Layout mode is better when approximate PDF coordinates matter, but it is less reflowable.
▸Can PDF scripts execute in the HTML?
No. PDF JavaScript is never extracted. The generated file contains no scripts and includes a restrictive content security policy.
▸Are fonts and images embedded?
Not in this first semantic exporter. Font attributes are approximated in layout mode; embedded fonts, arbitrary vector art, and most source images are disclosed as omitted.