Firecrawl AI features and pricing (October 2026)

Firecrawl is a web data API for developers who feed websites into AI applications. It scrapes, crawls, searches and maps sites and returns Markdown, HTML, screenshots or structured JSON.

Firecrawl offers endpoints for single pages (scrape), whole sites (crawl), site maps (map), web search, page interaction, monitoring and document parsing for PDF and DOCX files. SDKs exist for Python, Node.js, Go, Rust, Java, Elixir, Ruby, .NET and PHP, plus a CLI and REST API. The core is published on GitHub under the AGPL-3.0 license, so teams can also run it themselves. The operating company, SideGuide Technologies, is a Delaware corporation backed by Y Combinator, and the hosted service stores data in the United States.

• Extracts fields into a fixed schemaDevelopers pass a JSON schema to the scrape endpoint and receive the page content in exactly that shape, such as product name, price and availability. This removes hand-written parsers per website and gives downstream systems consistent records.
• Operates pages by instructionThe Interact endpoint clicks, scrolls, types and navigates on a page based on natural-language prompts. Data that only appears after an action, such as behind a filter or a load-more button, becomes reachable without writing browser scripts.
• Prepares web content for language modelsPages are returned as cleaned Markdown without navigation and boilerplate. Teams building retrieval or agent systems get shorter model inputs, which lowers token costs and keeps answers focused on the actual page content.

• Open-source core under AGPL-3.0 on GitHub alongside the hosted API, so self-hosting is possible.
• Credit-based pricing with a free plan of 1,000 credits per month and published plan prices.

Firecrawl is a self-service API sold worldwide; the operating company is a Delaware corporation that stores customer data on servers in the United States.

Check whether scraping the target sites is allowed by their terms and applicable law; Firecrawl respects robots.txt by default, but that does not settle the legal question. The hosted service stores data in the United States; zero data retention and SSO are limited to Enterprise. Estimate credit use per page type before choosing a plan, and note that AGPL-3.0 has obligations if you modify and offer the code as a service.

No reviews yet. Be the first to share how your team uses it.

Every review shows its verification level. We read each one before it goes live. How we check reviews