Tools / Convert PDF to JSON

Convert PDF to JSON

Convert PDF to structured JSON for apps and pipelines — sections, paragraphs, tables, images and page references. Not a single {"text": "..."} dump of the whole file.

Drop a PDF here

Choose a PDF

or drag and drop a file

Advanced options

OCR language

Limits: up to 100 pages · OCR up to 25 pages

No account required
Optional local OCR
Files up to 40 MB

What you get

Hierarchical document JSON (sections, paragraphs, tables, page refs) — not one big text string.

Structured JSON
Markdown (.md)
RAG chunks (.jsonl)
ZIP package

Limits: JSON reflects extracted structure; complex layouts may include review warnings.

How it works

Upload a PDF for structural analysis.
Sections, tables and images are mapped into a document model.
Download document.json for your pipeline.

Example output

Flat dump

{"text": "Introduction This manual covers installation... Chapter 2 Safety requirements..."}

Structured JSON

{
  "sections": [
    {"title": "Introduction", "page": 1, "paragraphs": ["This manual covers installation."]},
    {"title": "Safety", "page": 3, "paragraphs": ["Wear protective equipment."]}
  ]
}

Use cases

Import documents into custom apps
Build search indexes with page references
Automate downstream processing

Related tools

FAQ

What does the JSON contain?

Structured fields such as sections, paragraphs, tables, images and page_references — plus document metadata and warnings.

Is it just one big text string?

No. The goal is a hierarchical document model, not a single text field with the entire PDF.

Are page references included?

Yes, where available on sections and elements.

Can I also get Markdown?

Yes. Other formats remain available after conversion.

Does it work with scanned PDFs?

Yes. Scan pages are OCR’d when needed before structure is built.