Developer API Overview
Overview of the Smart PDF RAG API for parsing PDFs into structured JSON, Markdown, reading order, tables, page references, and bounding boxes.
What the API is for
The Developer API is designed for teams building PDF search, RAG, document review, extraction, and citation workflows. It focuses on structure that plain text extraction often loses.
Core output concepts
Useful output includes JSON elements, Markdown, headings, paragraphs, tables, reading order, page references, and bounding boxes where available.
- ✓Structured JSON
- ✓Markdown for ingestion
- ✓Page and bbox metadata
- ✓Tables and headings
- ✓Reading order
Use async workflows for real PDFs
Production PDFs can be large, scanned, or layout-heavy. An async task pattern lets applications submit a document, track processing, and retrieve results when complete.
Implementation checklist
- 1Create or select an API plan
- 2Authenticate requests
- 3Submit a PDF parse request
- 4Track the task result
- 5Store JSON and Markdown for downstream systems
Questions about Developer API Overview
Is the Developer API the same as PDF Accessibility?+
No. The Developer API is a separate product track for Smart PDF RAG and structured parsing workflows.
Why return both JSON and Markdown?+
Markdown is convenient for LLM ingestion, while JSON preserves typed elements, coordinates, tables, and metadata for review and citations.