Home / AI & ML / Firecrawl
Verified

Firecrawl

The open-source web data API that lets AI agents search, scrape and interact with the live web.

↗ firecrawl.dev Listed Oct 21, 2026 AI & ML · DevTools · Cloud & Infra

About Firecrawl

Firecrawl is a web data API for AI agents and LLM apps. Instead of running headless browsers and cleaning HTML yourself, you call one API to search the web, scrape a page, crawl a site, map a domain's URLs, or scrape a page and then interact with it using prompts or code. Output comes back as LLM-ready markdown, JSON or screenshots, with navigation, footers and ads stripped out.

It is aimed at developers building agents, RAG pipelines and data-collection workflows. Firecrawl is open source (its GitHub repository shows roughly 193K stars on the homepage), offers Python, Node.js, cURL and CLI access, and connects to AI agents and MCP clients with a single command. It also parses PDFs and DOCX files, waits intelligently for JavaScript-rendered content, and offers an autonomous Agent endpoint for gathering data from a plain-language request.

Pricing is credit-based and shared across endpoints: one credit buys a basic page scrape, crawl or map, and search costs 2 credits per 10 results. A free plan gives 1,000 credits a month with no card. Benchmark figures such as 96% coverage on the vendor's 1,000-URL dataset and the 93% token-saving claim are published by Firecrawl itself, so treat them as marketing claims rather than independent tests. The homepage also announces a $75M Series B.

Key features

Search
Search the web and get full page content from the results in one call.
Scrape and Crawl
Turn any URL or whole site into markdown, JSON or screenshots, including JavaScript-rendered pages.
Interact
Scrape a page, then click, type and navigate it with AI prompts or code.
Agent endpoint
Autonomous data gathering from a plain-language request, with five daily runs free.
Document parsing
Extract content from PDFs, DOCX and other files alongside web pages.
Open source
Core code is public on GitHub, so teams can inspect or contribute to it.

Frequently asked questions

What is Firecrawl?

Firecrawl is a web data API for AI agents. It searches the web, scrapes and crawls pages, maps a site's URLs and lets an agent interact with pages, returning clean markdown, JSON or screenshots ready for language models.

How much does Firecrawl cost?

There is a free plan with 1,000 credits a month. Hobby is $16 a month billed yearly ($19 monthly) for 5,000 credits, Standard is $83 a month billed yearly for 100,000 credits, Growth is $333 and Scale is $599 a month billed yearly. Enterprise is custom.

Is Firecrawl free?

Yes. The Free plan includes 1,000 credits per month, refreshed monthly, with no credit card required. That equals about 500 searches or 1,000 scraped pages.

How do Firecrawl credits work?

One credit covers a basic page scrape, crawl or map. Search costs 2 credits per 10 results, Interact costs 2 credits per browser minute, and JSON, Question and Highlight formats add 4 credits per page. Credits are shared across all endpoints.

Is Firecrawl open source?

Yes. The project is open source on GitHub at firecrawl/firecrawl, and the company also sells a hosted API with paid plans.

What are the alternatives to Firecrawl?

Common alternatives for giving agents web access include Tavily, Exa, Brave Search API and Serper. Tavily and Exa focus more on search, while Firecrawl is strongest at scraping and crawling full pages.

Reader endorsements 0 written

+ Write one

No endorsements written yet. Be first.

Related in AI & ML 3 picks

All topics
1.
O
Frontier AI models and the APIs to build on them.
AI & ML 4mos ago
—
2.
A
Claude — safety-focused AI models and API.
AI & ML 4mos ago
—
3.
H
The home of open-source machine learning.
AI & ML 4mos ago
—

Compare Firecrawl with…

Head-to-head with the competitors readers named.