Crawl4AI
Open-source crawler that turns web pages into LLM-ready Markdown
Overview
Crawl4AI is an open-source Python crawler that converts web pages into LLM-ready Markdown and structured data, with browser rendering, adaptive crawling, LLM extraction, a Docker server and MCP. Apache-2.0 with attribution; Crawl4AI Cloud (run from Singapore) gives $10 free credit, then packs from $10.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 1 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- pip install crawl4ai
- Self-host, no token
- 200 requests a day
- 2 concurrent requests
- 120 requests a minute
- 10 concurrent requests
- 1 credit = $0.001; page 0.2 credits
- Volume, regions and SLA
What you can produce with Crawl4AI
- Web pages to LLM-ready Markdown
- CSS, XPath and LLM-based structured extraction
- Browser rendering and adaptive crawling
- Docker server with JWT, dashboard and MCP
- Crawl4AI Cloud scrape, extract and search API
Crawl4AI is a practical, widely used open-source crawler for AI work, producing clean Markdown and structured data with browser rendering and LLM extraction, and its new hosted Cloud — run by a Singapore company — gives a cheap way to skip self-hosting. Its licence adds an attribution requirement to Apache 2.0, and the Docker server had several security fixes in 2026, so keep it updated. A strong free choice for developers building retrieval pipelines; use the Cloud credits to test before paying.
What we found
Crawl4AI is an open-source Python web crawler and scraper designed to turn web pages into clean, LLM-ready Markdown and structured data for AI applications such as retrieval-augmented generation and agents. It handles JavaScript-heavy sites through a real browser and offers adaptive crawling, CSS, XPath and LLM-based extraction, anti-bot handling, Shadow DOM support and a fast content-pruning filter, and it works with local models through Ollama via LiteLLM. A Docker server adds JWT authentication, a monitoring dashboard and an MCP server, and the library needs Python 3.10 or later. It was first published on GitHub in May 2024 by a developer known as unclecode, who is based in Singapore. Recent releases have focused on security: versions 0.9.3 in August and 0.9.4 in September 2026 closed eight disclosed vulnerabilities, including two server-side request forgery holes in the Docker server.
The library is licensed under Apache 2.0 with an added clause requiring attribution in distributions, which the project's README describes inconsistently as both recommended and required.
Crawl4AI Cloud, a hosted API with scrape, extract, search and answer endpoints, MCP access and batch jobs of up to 10,000 URLs, went live in September 2026. It is operated by Context4AI Pte Ltd, a Singapore company, under Singapore law, and processes data in the EU, Singapore and the US. During its soft launch new users get $10 of free credit, without a card, until 31 December 2026, with Community limits of 200 requests a day. Supporter credit packs cost $10, $25 or $100 and never expire, with one credit worth a tenth of a cent: a page costs 0.2 credits, and pages needing a real browser or harder sites cost more. A Developer subscription is coming, and Enterprise terms are by arrangement.
It suits developers who want a free, self-hostable crawler for AI pipelines, with a paid cloud as an option.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including Crawl4AI's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Crawl4AI unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Crawl4AI directly →
Spotted something out of date? Suggest an update →
More in Code & Dev Tools
