qwwiwi markdown.new -- Clean Web Content
Converts any webpage to clean Markdown. Reduces tokens by ~80% compared to raw HTML.
How to Use
Prepend https://markdown.new/ to any URL:
curl -sL "https://markdown.new/https://example.com/article"
When to Use
- Reading web pages, articles, blog posts
- Extracting documentation content
- Processing PDFs (built-in OCR)
- Getting clean text from any URL
When NOT to Use
- JavaScript-heavy SPAs (React/Vue apps)
- Login-protected content
- Real-time data (stock prices, live feeds)
- Pages that require interaction
Features
| Feature | Supported |
|---|---|
| Web pages | Yes |
| PDFs | Yes (OCR) |
| Images | Yes (description) |
| Audio | Yes (transcription) |
Integration Pattern
Instead of:
# Raw HTML -- wastes tokens
curl -sL "https://example.com/docs/guide"
Use:
# Clean Markdown -- 80% fewer tokens
curl -sL "https://markdown.new/https://example.com/docs/guide"
Setup
No signup required. Free service. No API key needed.
Fallback
If markdown.new is down, alternatives:
- Jina AI Reader -- prepend
https://r.jina.ai/ - Trafilatura -- self-hosted Python library