# Crawl4AI

> Crawl4AI (v0.9.5, October 2026) is a Python library, CLI and Docker server that drives a headless browser to crawl sites into Markdown with content filters and citations, and extracts JSON with CSS/XPath schemas (no LLM) or any LiteLLM provider, including Ollama. It has proxy, session and stealth options. Licence is Apache-2.0 with an added attribution requirement in the LICENSE file. A paid Crawl4AI Cloud adds web search and managed bot handling. Users must respect site terms and robots.txt.

- Repository: https://github.com/unclecode/crawl4ai
- Official docs: https://docs.crawl4ai.com
- Checked: 2026-10-16

## What it replaces
- Firecrawl Cloud: Crawl a site to Markdown, Scrape pages (JS-heavy and bot-protected), Structured data extraction (partial). You run the browsers and supply any proxies yourself; no built-in web search in the open-source version.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes, in your Python process or a Docker server on your machine | https://github.com/unclecode/crawl4ai#which-one | 2026-10-15 |
| Code licence | Partial | Apache-2.0 plus an added attribution requirement | https://github.com/unclecode/crawl4ai/blob/main/LICENSE | 2026-10-15 |
| Model weights licence | Confirmed | No separate model weights; LLM extraction uses the provider you connect | https://github.com/unclecode/crawl4ai#readme | 2026-10-15 |
| Commercial use | Partial | Yes, with the attribution required by the LICENSE file | https://github.com/unclecode/crawl4ai/blob/main/LICENSE | 2026-10-15 |
| Paid services needed? | Confirmed | None required; LLM extraction optional (local Ollama or paid APIs), Crawl4AI Cloud optional | https://github.com/unclecode/crawl4ai#which-one | 2026-10-15 |
| Hardware | Unknown | Not documented; runs a headless browser | https://github.com/unclecode/crawl4ai#readme | 2026-10-15 |
| Operating systems | Partial | Official installation guide includes Ubuntu-specific Playwright dependency instructions; Windows and macOS support are not confirmed in the reviewed installation pages | https://docs.crawl4ai.com/basic/installation/ | 2026-10-16 |
| Setup effort | Confirmed | Easy, pip install crawl4ai then crawl4ai-setup to install the browser | https://github.com/unclecode/crawl4ai#readme | 2026-10-15 |

GitHub: 85642 stars, 8896 forks, last push 2026-10-15, archived: no.
---
Source: RepoQuarry (https://repoquarry.dev/projects/crawl4ai). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.