AI GitHub
Marker for converting documents to structured text
Convert PDFs and supported files into Markdown, JSON or other documented formats.
Key Takeaways
- Marker uses document-processing models to identify and convert page elements.
- Choose Markdown for a readable inspection or a structured output when downstream code needs page or block information.
- Compare the Markdown and structured export for a scanned invoice and a digital PDF.
- Optional model-assisted processing changes resource and data-flow requirements.
How it works
Marker uses document-processing models to identify and convert page elements. Structured exports are useful when a downstream system needs more information than Markdown formatting carries.
What you need
Choose Markdown for a readable inspection or a structured output when downstream code needs page or block information. Make that choice before writing retrieval integration.
What to try first
Compare the Markdown and structured export for a scanned invoice and a digital PDF. Check table headers, reading order and how source pages are identified.
Include a long table broken across pages. Inspect whether headers and cells retain their association before feeding the result into a model.
Where to look for failures
Optional model-assisted processing changes resource and data-flow requirements. Review the code and model terms separately if you plan to use it commercially.
Related Franklin pages
Run the same scan through each approach and inspect the failed regions. An OCR engine and a full document-conversion pipeline can differ substantially in reading order and structural output.
- Detect text and analyze layout, reading order and tables in documents. Surya for OCR and document layout
- Detect and recognize text and extract structure with PaddleOCR pipelines. PaddleOCR for configurable document recognition
For how recognizing characters fits into the larger document processing workflow, see Franklin’s glossary. Optical character recognition
Project documentation
Try it for your workflow
Explore the project’s documentation for setup and usage, or save this page to a collection for later.
Read the project documentation ↗