Firecrawl
Firecrawl is a managed API platform that turns live web content into clean, AI-ready structured data for AI agents, RAG pipelines, and enterprise AI applications, serving 150,000+ companies including Apple, Shopify, Replit, and Zapier.
- Company typePrivate
- Founded2022
- HeadquartersSan Francisco, United States
- Headcount1–10
- GTM typeB2B
- OfferingSoftware
What Firecrawl does
Firecrawl is a managed web data platform purpose-built for AI applications, offering a single API surface that combines search, scraping, parsing, browser interaction, and autonomous research agents. Founded in 2022 in San Francisco by the team behind Mendable (Eric Ciarla, Nicolas Camara, and Caleb Peffer), the company exposes REST endpoints (/scrape, /crawl, /map, /search, /interact, /parse, /monitor, /agent, /deep-research) under a unified Fire-engine proprietary infrastructure that handles rotating proxies, headless browser rendering, smart-wait, caching, and indexing. The platform returns clean, token-efficient markdown or structured JSON, with specialized formats (Question, Highlights, Summary, Branding, Audio) and add-on modules including the Firecrawl MCP server, CLI, Skill bundle, Fire-PDF Rust parser, Browser Sandbox, Firecrawl Research Index (3M+ arXiv papers), and the proprietary Spark 1 model family powering /agent and Parallel Agents.
The business runs a primarily self-serve, developer-led PLG motion with a free keyless tier (1,000 credits/month) and paid subscriptions from $19/mo (Hobby) through $99/mo (Standard), $399/mo (Growth), and $599/mo (Scale), supplemented by usage-based credit overages and a quote-based Enterprise tier with SOC 2 Type II, ZDR, SSO, and dedicated SLAs. Distribution is amplified through native marketplace listings on Vercel, n8n, OpenRouter, and Lovable, plus an official Anthropic Claude Code plugin, MCP integrations for Cursor and Windsurf, and an open-source repo at 138.1K GitHub stars with 2.5M+ weekly SDK downloads. Reported adoption includes 1.25M+ developers, 150,000+ companies (including Apple, Shopify, Canva, Alibaba, DoorDash, Replit, Zapier, Gamma, You.com, Sprinklr, Cognism, and Sierra), and 5B+ requests served.
Revenue comes from tiered SaaS subscriptions and usage-based credit consumption, augmented by enterprise contracts and marketplace/partner-channel distribution. The company closed a $14.5M Series A in August 2025 led by Nexus Venture Partners (with Shopify CEO Tobias Lütke and Y Combinator) while reportedly already profitable, operating with a lean distributed team across the US, Brazil, and Europe.
Firecrawl firmographics
Firmographics- Name
- Firecrawl
- Legal name
- Firecrawl
- Website
- https://www.firecrawl.dev
- Company type
- Private
- Founded year
- 2022
- Operating status
- Operating
- Headcount range
- 1–10 employees
- Short description
- Firecrawl is a managed API platform that turns live web content into clean, AI-ready structured data for AI agents, RAG pipelines, and enterprise AI applications, serving 150,000+ companies including Apple, Shopify, Replit, and Zapier.
- Ownership category
- akta.pro rank
Where Firecrawl is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices3 records
Markets served
Firecrawl business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Infrastructure, Personnel, Marketing or Sales, Operations
Revenue model
- Tiered SaaS subscriptions (Free / Hobby / Standard / Growth / Scale): Monthly recurring subscriptions priced by monthly credit allotment: Free (1,000 credits), Hobby $19/mo (5,000), Standard $99/mo (100,000), Growth $399/mo (500,000), Scale $599/mo (1,000,000). Annual billing available with stated savings ($38/$198/$798/$1,798). Drives the majority of revenue through predictable recurring plans with auto-recharge upsell.
- Usage-based credit consumption and overages: API calls consume credits by endpoint: Scrape/Crawl/Map/Monitor 1 credit/page, Search 2 credits per 10 results, Interact 2 credits/browser minute, Agent (Preview) dynamic pricing with 5 free daily runs. Customers can buy add-on credit packs or enable auto-recharge for excess usage; enterprise customers get ZDR and SSO.
- Enterprise contracts (custom credits, SLAs, security): Enterprise tier offers unlimited scraping, custom concurrent requests, dedicated support & SLAs, bulk discounts, zero-data retention, SSO and advanced security — sold via a Typeform contact-sales flow. Pricing is quote-based.
- Marketplace- and partner-channel distribution fees: Firecrawl is distributed and billed through third-party marketplaces (Vercel Marketplace, n8n Cloud, Lovable connector, OpenRouter), where customers discover and pay through the partner — generating revenue share / referral economics alongside direct SaaS billing.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Monthly | Free Plan — 1,000 credits/month, 2 concurrent requests, low rate limits, no credit card |
| Subscription | Annual | Hobby — 5,000 credits/month, 5 concurrent requests, basic support |
| Subscription | Annual | Standard (Recommended) — 100,000 credits/month, 50 concurrent requests, standard support |
| Subscription | Annual | Growth — 500,000 credits/month, 100 concurrent requests, priority support |
| Subscription | Annual | Scale — 1,000,000 credits/month, 150 concurrent requests, priority support |
| Other | Multi-year contract | Enterprise — Custom credits, unlimited pages, custom concurrency, ZDR, SSO |
| Usage-based | Pay-as-you-go | Credit-based usage overage and add-on packs |
Go-to-market motion5 records
Distribution channels9 records
Marketing channels9 records
Firecrawl product offering
Product offeringCore offering
Firecrawl is a managed web data platform that exposes a single API to search, scrape, crawl, map, parse, monitor, and interact with live websites, turning dynamic, JavaScript-heavy pages into clean, LLM-ready markdown, JSON, screenshots, or structured data. The offering is built on a proprietary Fire-engine infrastructure and includes specialized endpoints (such as /agent, /interact, /monitor, and the Firecrawl Research Index) plus SDKs, a CLI, an MCP server, and self-host/open-source deployment.
Product overview
Firecrawl is a unified web data platform for AI applications, offered as a single API surface plus an ecosystem of specialized endpoints, modules, and integrations rather than a portfolio of separate products. The core platform (Firecrawl) exposes endpoints including /scrape, /crawl, /map, /search, /interact, /parse, /monitor, and /agent that work together to search, scrape, and act on the live web; Fire Engine is the proprietary hosted infrastructure underneath, while the open-source repo (130K+ GitHub stars) is also available for self-hosting. Layered on top are AI capabilities including the FIRE-1 Web Action Agent, the Spark 1 model family (Spark 1 Fast, Spark 1 Mini, Spark 1 Pro) powering /agent, and formats like Question, Highlights, Branding, and Summary. Add-on modules include Firecrawl MCP Server, Firecrawl CLI, Firecrawl Skill, the Firecrawl Browser Sandbox, Fire-PDF (Rust PDF parser), LLMstxt.new, Deep Research API, Lockdown Mode, and the Firecrawl Research Index (specialized AI/ML research index). The offering also extends through distribution channels — the Vercel Marketplace, n8n Cloud, OpenRouter, an official Claude plugin, Lovable connector, and 20+ other integrations — plus official SDKs for Python, Node.js, Go, Rust, Java, Elixir, Ruby, PHP (Laravel), and .NET.
Differentiator
Problem solved
Functional benefit
Products and services
- Firecrawl platform Unified web data API that lets developers search, scrape, crawl, map, interact with, parse, and monitor websites at scale, returning clean markdown, JSON, screenshots, or structured data for AI agents and RAG pipelines.
- /scrape endpoint Turns a URL into clean markdown, HTML, screenshots, or structured JSON via schema; handles JavaScript rendering and supports Question, Highlights, Summary, Branding, Audio, and JSON output formats.
- /crawl endpoint Recursively follows links from a starting URL and scrapes an entire site or section, with depth/page filters, subdomains, webhooks, and Smart Crawl prompts.
- /search endpoint Discovers web pages by query and returns results with optional full-page markdown content; supports web/news/images sources and categories (github, research, pdf), with language, country, and time-range parameters.
- /interact endpoint Browser automation endpoint that scrapes a page and then performs clicks, form fills, navigation, and data extraction via natural-language prompts or Playwright/Bash code, with reusable browser profiles and live view URLs.
- /map endpoint Quickly returns the URLs of a website for URL discovery and site mapping, supporting up to 100,000 results per request.
- /agent endpoint Autonomous web research agent that searches, navigates, and gathers data from complex websites to find information in hard-to-reach places, powered by the proprietary Spark 1 model family with Parallel Agents support.
- /parse endpoint Turns documents (PDF, DOCX, DOC, ODT, RTF, XLSX, XLS, HTML up to 50 MB) into clean markdown, JSON, or summaries, powered by the Rust-based Fire-PDF engine with Zero Data Retention on Enterprise plans.
- /monitor endpoint Watches pages for changes and notifies AI agents via signed webhook or email when content changes, with goals set in plain English, configurable cadence from 5 minutes to custom cron, and permalinks for each change diff.
- Firecrawl MCP Server Official Model Context Protocol server that gives AI agents (Claude Code, Cursor, Windsurf, Codex) access to live web data — scrape, search, crawl, and map — with FIRE-1 interaction support and SSE streaming, hosted at mcp.firecrawl.dev/v2/mcp.
- Firecrawl CLI Command-line interface providing full scrape, search, crawl, and map support, installable via npm or a one-shot npx command, designed for terminal workflows and agent-driven sessions.
- Firecrawl Skill Three-skill agent package (CLI skills, build skills, workflow skills) that teaches AI agents how to drive Firecrawl during a session, integrate Firecrawl into application code, and produce repeatable deliverables such as research briefs, SEO audits, lead lists, and QA reports.
- Firecrawl Research Index
Quantifiable outcome
- 94% fewer input tokens per page (35,980 tokens saved on a typical page); 28.4x+ ROI on Hobby plan against Claude Sonnet 4.6 inference cost
- +10 more outcomes
Companies that use Firecrawl
Customer profileNamed customers31 records
Segments5 records
Ideal customer profiles4 records
Firecrawl technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration13 records
AI capability14 records
Feature8 records
Firecrawl partnerships and signals
Strategic signalPartnerships
Ten partnerships are on record, tiered core and flagship.
- Vercel (Marketplace)coreFirecrawl is a native integration on the Vercel Marketplace: one-click install, API key auto-injected into the project environment, and billing handled through Vercel.
- OpenRoutercoreFirecrawl /search is available as a web search engine on OpenRouter so any model on OpenRouter can ground with Firecrawl's live, full-page, markdown-ready web content via a single toggle.
- n8ncoreFirecrawl is natively integrated into n8n Cloud: one-step connect with no API keys needed and 10,000 free credits when users connect through n8n.
- Wikimedia EnterpriseflagshipFirecrawl routes all Wikipedia requests through Wikimedia Enterprise's paid On-demand API, compensating Wikimedia directly for high-volume programmatic access. Improves speed (no headless browser), consistency, and attribution; part of Firecrawl's broader effort to launch a Creator Program to compensate content creators.
- Anthropic (Claude Code plugin)flagshipFirecrawl is an official Claude Code plugin (announced Feb 2026) giving Claude direct access to live web data — scrape pages, crawl sites, search the web, and extract structured data without leaving the workflow.
- LovableflagshipFirecrawl integrated directly into Lovable's connector system so users can create web-data apps by describing them in plain English; users generate Firecrawl API keys inside Lovable and get 10,000 free credits on connect (special credit offer).
- Cursor / Windsurf (MCP integrations)flagshipOfficial Firecrawl MCP server integrates with Cursor, Windsurf, Claude Code and other MCP-compatible editors; one command (`npx -y firecrawl-mcp`) installs the server with over 400,000 MCP servers installed to date.
- StripecoreFirecrawl accepts payments through Stripe, supporting most major credit cards, debit cards, and PayPal for all subscription tiers.
- WorkOScoreFirecrawl supports WorkOS ID-JAG identity assertions for agent platforms that can mint them, allowing direct authentication to Firecrawl without browser/CLI auth flows.
- GitHubcoreFirecrawl offers a free remote MCP server for GitHub, listed among seven free remote MCPs developers can use to manage planning, design, coding, and research workflows.
Scale indicators15 records
Recent moves6 records
Expansion highlights6 records
Firecrawl competitors and assessment
Company assessmentDirect peers
- Apify: Apify is a full-stack web scraping and automation platform with a marketplace of "actors" for scraping any site. It competes with Firecrawl for developers building scrapers and agents, and is benchmarked head-to-head (e.g., Alex Reibman's AgentOps test that showed Firecrawl 50x faster than Apify).
- ScrapingBee: ScrapingBee provides a web scraping API focused on handling headless browsers, proxies, and CAPTCHAs, returning rendered HTML for downstream processing. It is a comparable developer-facing scraping API with similar target users and a similar pricing model.
- Exa (formerly Metaphor): Exa offers neural/semantic web search APIs aimed at AI applications, returning ranked results with full content for agent workflows. It overlaps directly with Firecrawl's /search and /scrape for AI-grounded retrieval and is named in Firecrawl's own comparison pages.
- Tavily: Tavily is an AI-native search API built specifically for LLM agents and RAG pipelines, offering search, extract, and crawl endpoints with clean markdown output. It is the most directly comparable AI-first alternative to Firecrawl's /search, /scrape, and /crawl endpoints and competes for the same agentic AI buyer.
Broad incumbents
- Zyte (formerly Scrapinghub): Zyte is an established enterprise web data extraction platform offering Scrapy Cloud, AI-powered extraction, and managed scraping services. It serves large enterprise data programs at greater scale than Firecrawl and competes on the same broad use cases.
- Bright Data: Bright Data is one of the largest commercial web data platforms, offering proxies, scraping APIs, and pre-collected datasets at enterprise scale. It is a broader incumbent with overlap on data collection and scraping infrastructure that competes for enterprise web data workloads.
Emerging players
- Browser Use: Browser Use is an emerging open-source framework that lets AI agents control a browser to navigate, click, and extract data from any site. It competes directly with Firecrawl's /interact and Browser Sandbox offerings in the agentic browser-control layer.
- Crawl4AI: Crawl4AI is an open-source, AI-friendly web crawler/scraper optimized for LLM data ingestion. It is a free, self-hostable alternative to Firecrawl's managed scraper with overlapping feature set and a strong GitHub community.
- Parallel: Parallel offers AI-native web search and research APIs (including Parallel Search Pro) aimed at agent workloads. It was named alongside Firecrawl in the AIMultiple top 4 of agentic search APIs and competes for the same AI search API budget.
- Jina AI: Jina AI offers Reader, Embeddings, and Search APIs focused on converting web content into LLM-ready formats, with strong open-source traction. It overlaps meaningfully with Firecrawl's URL-to-clean-data proposition, especially for RAG developers.
Market position
Strengths1 record
Weaknesses1 record
Competitive moat6 records
Key risks5 records
Key highlights7 records
Customer concentration
Firecrawl social profiles
Digital presenceFirecrawl compliance and trust
Trust signalCompliance1 record
Firecrawl financial estimates
Financial estimateRevenue estimate
Valuation estimate
Firecrawl leadership team
Management profileNumber of profiles
Profiles3 records
Firecrawl subsidiaries and ownership
Company hierarchySubsidiaries1 record
Firecrawl funding detail
Funding detailFunding overview
Funding rounds4 records
Investors7 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Firecrawl M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Firecrawl
What does Firecrawl do?
Firecrawl is a managed web data platform that exposes a single API to search, scrape, crawl, map, parse, monitor, and interact with live websites, turning dynamic, JavaScript-heavy pages into clean, LLM-ready markdown, JSON, screenshots, or structured data. The offering is built on a proprietary Fire-engine infrastructure and includes specialized endpoints (such as /agent, /interact, /monitor, and the Firecrawl Research Index) plus SDKs, a CLI, an MCP server, and self-host/open-source deployment.
Is Firecrawl a public or private company?
Firecrawl is a private company. It is classified as venture growth investor backed and is currently operating.
When was Firecrawl founded?
Firecrawl was founded in 2022. It employs 1 to 10 people.
Where is Firecrawl based?
Firecrawl is headquartered in San Francisco, United States, in the North America region.
How does Firecrawl make money?
Four revenue lines are on record. Tiered SaaS subscriptions (Free / Hobby / Standard / Growth / Scale) is the primary driver. The others are usage-based credit consumption and overages, enterprise contracts (custom credits, SLAs, security) and marketplace- and partner-channel distribution fees.
Who are Firecrawl's main competitors?
Direct peers on record are Apify, ScrapingBee, Exa (formerly Metaphor) and Tavily. Broad incumbents are Zyte (formerly Scrapinghub) and Bright Data. Emerging players are Browser Use, Crawl4AI, Parallel and Jina AI.
Does Firecrawl have an API?
Yes. Firecrawl offers a public REST API (base URL https://api.firecrawl.dev/v2, Bearer auth with FIRECRAWL_API_KEY) exposing endpoints including POST /search, /scrape, /interact, /crawl, /map, /parse, /monitor, /agent, /deep-research, and support endpoints /support/ask and /support/docs-search. The API returns clean markdown by default, with optional formats including raw HTML, screenshots, JSON via schema, summary, question (grounded answers), and highlights. SDKs are offered for Python, Node.js, Go, Rust, Java, Elixir, Ruby, PHP (with Laravel support), and .NET. Official Firecrawl MCP server (https://mcp.firecrawl.dev/v2/mcp) is available and supports HTTP Transport and SSE modes; over 400,000 MCP servers have been installed. A CLI (`npx -y firecrawl-cli@latest`) and Skill packages are also offered for terminal-based and agent-based workflows. Rate limits are tier-based (concurrent browsers per plan), and a keyless free tier with limited rate caps is offered as a fallback. Authentication also supports WorkOS ID-JAG identity assertion for supported agent platforms. Developer documentation is at docs.firecrawl.dev.