Firecrawl Open-Sources anydoc: Convert 14 Document Formats to Markdown in Milliseconds
aigclink · x · 2026-08-12
Firecrawl has released anydoc, a Rust-based document parsing library that converts formats like Word, PPT, Excel, and PDF into clean GitHub-Flavored Markdown.
Key highlights:
- Blazing fast: Parses a single file in under 5 milliseconds, processing 500 DOCX files in just 1.7 seconds.
- Unified output: All formats pass through a single model, ensuring consistent behavior for table escaping, heading anchors, and footnotes.
- Extension-agnostic: Identifies formats based on content rather than file extensions, correctly handling misnamed files.
- Multi-platform: Includes bindings for Node.js, Python, and WebAssembly for in-browser execution.
Related event: Firecrawl Open-Sources anydoc for Millisecond Document Parsing(3 posts)→
More from coding & agent
- AI Discovers New WiFi Hacking Method After Getting Root Access Overnight — evilsocket · 2026-08-12
- MCP Increases Token Costs: Developer Reveals the Price of Multi-Turn Interactions — KitchenAmoeba4438 · 2026-08-12
- Anthropic's Interactive Prompt Engineering Tutorial Hits 40k Stars — thisguyknowsai · 2026-08-12
- Anthropic Academy Launches with Free Official Prompt Engineering Courses — thisguyknowsai · 2026-08-12
- From Prompt to Harness Engineering: The 3 Stages of AI Agent Architecture — femke_plantinga · 2026-08-12
- Firecrawl Open-Sources pdf-inspector: Extracts 200 PDFs in 0.47s — aigclink · 2026-08-12