Fire-PDF Overview
What Fire-PDF is, where the speed comes from, and how it preserves structure like tables, formulas, and multi-column layouts.
Fire-PDF Documentation
Structured output, correct reading order, and GPU only when needed.
What you’ll find here
Section titled “What you’ll find here”Document Parsing
Supported formats, PDF parsing modes, usage examples, and output format.
Launch Article
A narrative write-up (currently available in Chinese) covering why Fire-PDF exists, how the pipeline works, and what changed vs. the previous parser.
Recommended paths
Section titled “Recommended paths”Quickstart
From a PDF URL to structured Markdown with minimal configuration.
PDF Parsing Modes
How to choose fast/auto/ocr based on PDF types and goals.
Errors & Troubleshooting
Turn failure cases into reusable strategies with sampling and routing.
Performance & Cost
Control cost with maxPages and a two-stage parsing approach.
FAQ
Quick answers for missing body text, scanned PDFs, and broken tables.