Crawl4AI Online: turn any web page into LLM-ready Markdown

What is Crawl4AI?

Crawl4AI is an open-source Python crawler that turns websites into clean Markdown and structured data for large language models. It is developed by unclecode on GitHub, released under the Apache 2.0 license and runs a real headless browser (Playwright), so it also handles pages that load their content with JavaScript. Typical uses are RAG pipelines, AI agents, research datasets and price or content monitoring.

The library itself has no web interface. You install it with pip or run the official Docker image. This page gives you the result without that setup – and hands you the code when you want more.

How the online tool works

1. Paste a URL. Any public web page: article, documentation, product page, blog post.
2. Choose options. “Main content only” removes navigation, cookie banners, sidebars and footers – the same idea as Crawl4AI’s fit_markdown. Keep or drop links and images.
3. Copy or download. You get Markdown with headings, lists, tables, code blocks and absolute links, plus word and token counts for your context window.
4. Take the code. The Python · Crawl4AI tab shows a ready-to-run script with exactly your options. The Docker API tab shows the same request against the official Docker server.

What you can do with the Markdown

Crawl4AI in 30 seconds (Python)

pip install -U crawl4ai
crawl4ai-setup        # installs the Playwright browser
crawl4ai-doctor       # optional: checks the installation

import asyncio
from crawl4ai import AsyncWebCrawler

async def main():
    async with AsyncWebCrawler() as crawler:
        result = await crawler.arun(url="https://example.com")
        print(result.markdown)

asyncio.run(main())

Step-by-step guide with filters, multiple URLs, deep crawling and LLM extraction: Crawl4AI Python tutorial. Prefer an API? See Crawl4AI Docker. Using Claude or Cursor? See the Crawl4AI MCP server.