PDF Extraction
Upload an existing PDF and extract its content back into editable LaTeX source code.
How it works
- Go to PDFs in the dashboard sidebar
- Upload your file, which supports up to 50 MB and 100 pages
- Click Extract LaTeX: the engine parses document structure, headings, lists, equations, and figures
- Review the extracted
.texfile in the preview - Download it directly, or import into a new project with one click
What gets extracted
- Document structure: sections, subsections, paragraphs
- Math: inline and display equations, aligned environments
- Lists: ordered and unordered
- Tables: basic table structures
- Figures: placeholder references for images
Best practices
- LaTeX-generated PDFs produce the most accurate results since the structure is already well-defined
- Word-generated PDFs work well for text and basic formatting
- Scanned PDFs go through an additional OCR step, so results depend on scan quality
- Multi-column layouts may need manual cleanup after extraction
Note: Extraction uses your monthly page quota. The free tier includes 40 pages per month. See Pricing for higher limits.
Using the API
Automate PDF extraction via the REST API. See PDFs API for endpoints and examples.