
Firecrawl is a powerful, open-source web data API that turns web pages into clean, LLM-ready data for AI applications. Firecrawl handles crawling, scraping, and data extraction at scale.
Key Features:
Whether you're building AI applications, training LLMs, or powering RAG pipelines, Firecrawl provides the web data infrastructure you need.
Firecrawl fits AI teams that want one API for scraping, crawling, search, screenshots, and structured extraction. The main caveat is its AGPL-licensed core if you need permissive embedding.
Growing a website? Track its keywords, Google rankings, Search Console insights, and share of voice with bisibility. Free during beta.
Our productSource code
firecrawl/firecrawl
Stars
191.9KForks
10.1KLast commit
2026-10-15License
AGPL-3.0-or-laterLanguage
TypeScriptMetrics recorded 2026-10-16
License source: Repository license file
Firecrawl provides APIs to search, scrape and crawl web content, returning Markdown or structured data for applications and AI workflows. Its Cloud service manages the fetching infrastructure; the open-source version lets you operate the core API stack yourself.
Sources (checked ): Firecrawl introduction; Open source and Cloud
Yes. Firecrawl documents a Docker Compose setup for its open-source stack. Self-hosting gives you responsibility for infrastructure, security, storage and upgrades. The default stack does not include every Cloud capability, so review the edition comparison before relying on feature parity.
Sources (checked ): Firecrawl self-hosting; Edition comparison
Firecrawl Cloud’s published Free plan offers 1,000 credits per month. Endpoints and advanced features consume credits at different rates, so that allowance is not a guarantee of 1,000 arbitrary operations. Self-hosting instead brings infrastructure and any external-provider costs.
Sources (checked ): Firecrawl Free plan and credits; Self-hosted cost responsibilities
Cloud usage is constrained by credits, concurrency and plan limits. The open-source default stack has different capabilities from Cloud, including limits around managed anti-bot services and interactive features. Neither the edition comparison nor the credit model establishes guaranteed access to every protected page.
Sources (checked ): Firecrawl plan limits; Open-source capability boundaries
Crawl4AI offers an embeddable Python crawler as well as a hosted Cloud service. Firecrawl exposes its scraping and crawling capabilities through an API, available through Cloud or a self-hosted stack. Compare the Python-library or API integration you need and each deployment’s documented feature set.
Sources (checked ): Crawl4AI Python and Cloud; Firecrawl deployment options
Verified against the firecrawl/firecrawl repository and the npm package.
Compiled from public repository, package, and project documentation data; check the vendor's site for current details.
Platform or service · SaaS / API
Platform or service · API
+1 more
Platform or service · API