PDF to JSON
Turn a PDF from a public URL into structured JSON — typed blocks with text, page, and bounding box per element. Run it in your browser with a free AgentBody account.
- Real Structured JSON
- Element Types Preserved
- Page-Aware Output
- Works From A URL
- Markdown Included
- Agent-Ready Workflow
Input
Output
Run the tool to see the result here.
Features
What this PDF to JSON converter does and how it fits data extraction, pipelines, and LLM workflows.
Real Structured JSON
The conversion returns typed blocks — each carrying its id, element type, text content, page number, and bounding box. That is machine-readable structure, not a flat text dump: fields your code can index, filter, and store directly.
Element Types Preserved
Blocks carry a type field distinguishing headings, paragraphs, and other document elements, so your downstream logic can treat a title differently from body text without writing a parser.
Page-Aware Output
Every block knows its page number, and pages come with their own text layer. Page-level and element-level JSON sit side by side in one response — slice by page, or work block by block.
Works From A URL
Pass the PDF's public HTTPS URL — a report, an invoice, a paper hosted anywhere public. The file_url plus a file_name is all the request needs; no file picker, no upload step.
Markdown Included
The same result carries a markdown rendering of the document, so you get clean prose for reading and LLM prompts alongside the structured blocks for programmatic use — one parse, both shapes.
Agent-Ready Workflow
This page calls the same POST /v1/documents/parse endpoint your agents use, and the stored result reads back through GET /v1/documents/{document_id} — manual runs here, API automation later, identical data.
How to use
Follow these steps to convert a PDF to JSON: prepare the URL, run the parse, and read the structured blocks.
Create A Free Account
Sign in or create a free AgentBody account. The account unlocks the conversion and carries the credit cost.
Copy The PDF URL
Open the PDF you want to convert and copy its public HTTPS URL and file name. Signed, expiring, or intranet links will not resolve.
Fill The Form
Enter the file_url and file_name on this page. Optionally enable merge_tables or relevel_titles to tune how tables and headings come through.
Run The Parse
Click Run. The request goes to POST /v1/documents/parse and returns a document_id — the structured result is stored against your account.
Read The JSON
Open the stored result: blocks with id, type, text, page, and bbox arrive alongside per-page text — queryable by page range through the read-back endpoint.
Use The Result
Feed the blocks array into your database, search index, or LLM pipeline. The fields are stable across requests, so your extraction code writes once and keeps working.
Frequently asked questions
How do I convert a PDF to JSON?
Copy the PDF's public HTTPS URL and file name, paste them into the form on this page, and click Run. The parse stores a structured result you can read back as JSON — typed blocks with text, page numbers, and positions, plus per-page text.
Is this a free PDF to JSON converter?
You can start free: the tool runs in your browser with a free AgentBody account, and each parse uses your account's credits. Pricing is governed by the gateway and shown in the console — not fixed numbers on this page.
What JSON structure does the converter return?
A blocks array where each element carries id, type, text, page, and bbox (its position on the page), plus a pages array with per-page text and a markdown field. Structure at the element level and the page level in one response.
How do I turn a PDF book into JSON format?
Parse it once with the public URL, then read the stored result back page by page — page_start and page_end parameters let you walk a long book through in ranges instead of one huge response, assembling the full JSON at your own pace.
Is there a PDF to JSON converter free of upload steps?
There is no upload at all — you point this converter at a public HTTPS URL and it fetches the document itself. That makes it a PDF to JSON converter free of file-picker friction: URL in, structured blocks out, with each parse metered on your account's credits.
How do I turn a PDF book into JSON format page by page?
Parse the book once from its public URL, then read the stored result back with page_start and page_end ranges. Every block already carries its page number, so chapters and sections assemble into JSON at whatever granularity your project needs.
Does it convert scanned PDFs?
The parser reads documents with an extractable text layer. A pure scan without one will not produce usable blocks — no OCR promise is made here. Text-based PDFs convert directly.
Can it convert JSON back to PDF?
No — this tool runs one direction, PDF to JSON. Generating a PDF from data is a different job; your own rendering pipeline can consume the JSON this tool produces.