Firecrawl's pdf-inspector: Auto-Detect PDF Type, Extract to Markdown in 200ms

Firecrawl open-sourced pdf-inspector, a tool that automatically determines whether a PDF is text-based or scanned. For text-based PDFs, it extracts text directly and converts to clean Markdown, with table detection, multi-column layout recognition, and heading hierarchy restoration, achieving speeds as fast as 200ms. It provides Python, Node.js, Rust, and browser (WebAssembly) interfaces, enabling use without a backend. The tool is suitable for PDF pre-classification and extraction in document processing pipelines.