I built pdf2markdown.io – both an online tool and an API for converting PDFs and images to Markdown.
- Online tool: Upload and convert, no signup required for quick tries
- Image-based PDFs: works with scanned docs and image-only PDFs (no extra OCR needed)
- Images: JPG, PNG, etc. → Markdown
Useful for RAG pipelines, document extraction, or quick one-off conversions.
Anyone else here building AI agents that need web data?
So I built *AnyCrawl* specifically for AI use cases:
Handles JS-heavy sites (React, Vue, etc.)
20x performance boost through multi-processing
Support proxies (no more blocks!)
Returns LLM-friendly structured data
Open source (MIT) + optional SaaS
- Online tool: Upload and convert, no signup required for quick tries - Image-based PDFs: works with scanned docs and image-only PDFs (no extra OCR needed) - Images: JPG, PNG, etc. → Markdown
Useful for RAG pipelines, document extraction, or quick one-off conversions.