Introducing the Government & Legal Index
Give your AI agents statutes, regulations, local codes and court opinions from 30 million US government pages, free to search in ChatGPT, Claude and the API.
Changelog
The Company Enrichment Benchmark
We tested the leading company enrichment providers on hundreds of U.S. companies to see who delivers accurate data and where information is missing.
Read the full updateFree People Data for ChatGPT and Claude
Give your AI agents access to 270M+ company and people records in Alexandria, with 1,000 free credits a month through the Firecrawl plugin.
Read the full updateFirecrawl for Google Cloud
We're partnering with Google Cloud to give Gemini and ADK agents data from the web and beyond, available on Google Cloud Marketplace.
Read the full updateIntroducing Fireround
Fireround is a new company funding data provider built by Firecrawl. Follow new rounds and research investors in ChatGPT and Claude through Alexandria.
Read the full updateIntroducing the Marketing Data Pack in Alexandria
Uncover what gets traffic and attention, shape your SEO and GEO strategy, and create content and visuals with better references in ChatGPT or Claude.
Read the full updateIntroducing Universal Scrape
Web pages, files and a growing set of Alexandria providers, all with one Firecrawl API.
Read the full updateFantasy sports data is live in Alexandria
Use fantasy rankings, projections, sports betting odds and live news in ChatGPT or Claude to research your next lineup with Firecrawl.
Read the full updatePeople and company enrichment in Alexandria
Apollo, FullEnrich, and Data Legion give your agents people search, work email lookup, and company enrichment through Firecrawl.
Read the full updateAlexandria and the Developer Index are now in ChatGPT and Codex
Use the Firecrawl plugin to search and scrape the web, access 100+ Alexandria data providers, and research code with the Developer Index in ChatGPT and Codex.
Read the full updateAlexandria is now in Claude and Claude Code
Give Claude access to 100+ data providers and Firecrawl's specialized indexes alongside live web search and scraping.
Read the full updateIntroducing Alexandria and our $75M Series B
Meet Alexandria, the knowledge library for superintelligence, and our $75M Series B to invest in search and broader access to knowledge.
Read the full updateFirecrawl Developer Index
70M+ primary sources across GitHub repos, docs, issues, and PRs, with the highest recall of any major coding-specific index.
Read the full updateLife Sciences in Firecrawl Research Index
41M+ papers across the drug discovery, clinical trial, and biology literature at 90% recall@10, and the whole index is free to use.
Read the full updateIntroducing our most accurate /search yet
A new custom relevance model returns the excerpts that best answer your query. Agents using Firecrawl /search score 94.7% on SimpleQA, higher than with any other provider tested, while using 10x fewer tokens.
Read the full updateWeb-scale /monitor
Always-on search across the entire web: define a query and goal, and get pinged or webhooked the moment something relevant comes online.
Read the full updatev2.11.0 is live
Firecrawl v2.11.0 ships the Firecrawl Research Index, keyless endpoint access, automatic PII redaction, a deterministicJson format, and video discovery on any page - plus much more.
Read the full updateFirecrawl Research Index
A specialized research index for AI agents: 3M+ arXiv papers and the GitHub code behind them, with state-of-the-art recall on arXivQA at comparable cost.
Read the full updateIntroducing /monitor
Notify your AI agent via webhook the moment a page or site changes, and use up to 90% fewer LLM tokens by only ingesting what actually changes.
Read the full updatev2.10 is live
Firecrawl v2.10 ships the /parse endpoint, Lockdown Mode, Question and Highlights formats, and four new official SDKs.
Read the full updateHighlights Format
A new format for /scrape that returns the exact sentences, code blocks, and table rows on a page matching your query, verbatim, with up to 100x fewer tokens than a full scrape.
Read the full updateQuestion Format
Ask any web page a question and get a grounded answer with zero hallucinations, using up to 100x fewer tokens than returning the whole page.
Read the full updateLockdown Mode
Serve /scrape results exclusively from Firecrawl's existing index with no outbound traffic and zero data retention by default.
Read the full updateIntroducing /parse
Upload PDFs, Word docs, and spreadsheets, and /parse turns them into clean Markdown, JSON, or summaries for AI agents and RAG pipelines.
Read the full updateIntroducing web-agent
Firecrawl `web-agent` is an open framework for building AI agents that search, scrape, and interact with the web. Powered by the same architecture behind our `/agent` endpoint.
Read the full updateIntroducing Fire-PDF
Fire-PDF is Firecrawl's new PDF parsing engine: up to 5x faster, smarter about what hits GPU, and more accurate on complex layouts.
Read the full updatev2.9.0 is live
Browser interaction via /interact, question and audio scrape formats, onlyCleanContent, PDF parsing modes, Java and Elixir SDKs, and dozens of fixes.
Read the full updateIntroducing /interact - Scrape and Automate Actions With Agents
Scrape any page, then let your agent click, type, and navigate in natural language or with full Playwright control.
Read the full updateFirecrawl Java SDK
Full support for our core endpoints, including scrape, search, and crawl. Works with Maven, Gradle, Java 17+.
Read the full updateFaster, Smarter PDF Parsing
Our new PDF parsing engine delivers 3x faster parsing and significantly improved reliability, rebuilt in Rust.
Read the full updateIntroducing Browser Sandbox
Firecrawl Browser Sandbox gives your agents a secure, fully managed browser environment
Read the full updateImproved Logo Extraction for Branding Format v2
We significantly improved logo extraction accuracy for Branding Format v2.
Read the full updatev2.8.0 is live
Major improvements to agent workflows, developer tooling, and self-hosted deployments across the API and SDKs.
Read the full updateParallel Agents: Extract Web Data at Scale
Batch process hundreds or thousands of /agent queries simultaneously with real-time streaming results.
Read the full updateIntroducing the Firecrawl Skill & CLI
A new way for AI agents to reliably access real-time web data with a single install.
Read the full updatev2.7.0 is here!
Introducing ZDR Search support, Partner Integrations API, Improved Branding Format and More
Read the full updatev2.6.0 available now
Introducing Unified Billing, Reliability and Speed Improvements, Branding Format, and More
Read the full updatev2.5.0 - The World's Best Web Data API
We now have the highest quality and most comprehensive web data available
Read the full updatev2.4.0 is live now
New PDF search category, 10x better semantic crawling, x402 search endpoint, and open-source upgrades!
Read the full updateFirecrawl v2.3.0 is here
Introducing YouTube transcript support, odt & rtf parsing, 50x faster docx parsing, and more
Read the full updateFirecrawl v2.2.0 is here
Introducing MCP Version 3, Webhooks Signatures, 15x faster map, and more
Read the full updateFirecrawl v2.1.0 is here
Introducing Search Categories, Image Extraction, Data Attribute Scraping, Hash-Based Routing and more
Read the full updateIntroducing Firecrawl v2
Firecrawl v2 ships faster scraping by default, a summary format, news and image search, and smart crawling with natural-language prompts.
Read the full updateFirecrawl v1.15.0 is here
Introducing SSO for enterprise, improved scrape reliability, FireGEO example and more
Read the full updateFirecrawl v1.14.0 is here
Introducing authenticated scraping, zero data retention for enterprise, improved speeds, and more
Read the full updateFirecrawl v1.13.0 is here
Introducing AU, FR, DE locations to stealth mode, crawl subdomains, Google Slides scraping, and more
Read the full updateFirecrawl v1.12.0 is here
Introducing Max Concurrency per request, Crawl Entire Domain, Google Docs Scraping, and more
Read the full updateFirecrawl v1.11.0 is here
Introducing Firecrawl Index for 5x faster scraping, improved activity logs, and more
Read the full updateIntroducing Search
Search the web and get full content from results in one API call
Read the full updateFirecrawl v1.9.0 Release
Self-host improvements, MCP enhancements, SDK updates, and performance boosts with increased limits.
Read the full updateEnhanced Mode pricing and quality updates
Enhanced Mode, the proxy option for hard-to-scrape sites, gets a quality upgrade and a pricing change, with guidance on using it as a retry step.
Read the full updateFirecrawl Launch Week III – Summary
Major updates including FIRE-1 agent, Python SDK v2.0, LLMstxt.new, and 20+ new integrations.
Read the full updatev1.7.0: Deep Research alpha, concurrent browsers
Deep Research open alpha with structured outputs, concurrent browsers raise rate limits, /extract no longer needs URLs, and crawl gets a depth option.
Read the full updateConcurrent browsers and 5x higher rate limits
Rate limits are now expressed as concurrent browsers and raised 5x across plans, with upgrades available for faster throughput.
Read the full updatev1.6.0: LLMs.txt API and the MCP server
Turn any site into llms.txt with /llmstxt, join the Deep Research alpha, and give Cursor, Windsurf and Claude web access through the Firecrawl MCP server.
Read the full updateSelf-host overhaul: v1.5.0
A reworked self-hosting guide, Kubernetes examples, proxy support, SearXNG search and a production Docker image, plus Deep Research alpha and llms.txt.
Read the full updatev1.4.4: PDF and image sub-link extraction
Extraction now reads PDF and image sub-links with Gemini, scrape validates actions and wait times, and new Gemini crawler and TrendFinder examples ship.
Read the full updateExamples week: v1.4.3
Open Deep Research, an R1 web extractor, an o3-mini crawler and more examples, with URL deduplication, better blocklists and scrape options for extract.
Read the full updateExtract and API improvements: v1.4.2
Web search inside extract with enableWebSearch, source tracking, configurable webhook events, a map timeout, and ad blocking on by default.
Read the full updateExtract improvements: v1.4.1
Extract returns more data, parsing moves from Cheerio to a Rust-based parser, infrastructure is more reliable, and crawl cancellation is more responsive.
Read the full update/extract is now asynchronous
Extract requests return a job ID to poll for status. The API stays backwards compatible for 10 days, and the SDKs handle the change for you.
Read the full updatev1.2.0: Introducing /search
The new /search endpoint returns web results for any query, with optional full page markdown via scrapeOptions, plus extract fixes and MinerU for PDFs.
Read the full updatev1.1.0: geolocation and mobile scraping
Geolocation and mobile scraping, 4x faster parsing, better webhooks, credit packs with auto-recharge, iframe support and URL deduplication, plus SDK fixes.
Read the full updateIntroducing Batch Scrape
Scrape many URLs in one request with the new Batch Scrape endpoint, now supported in the Python SDK 1.4 and the Node SDK 1.7.
Read the full updateCancel crawls from the SDKs and faster scrapes
Crawl cancellation lands in the Python and Node SDKs, new OpenAI Voice and CRM enrichment examples, improved Docker images and faster scrapes.
Read the full updateCrawl reliability fixes
Fixes for 500 errors on busy sites, crawl status for pages over 10 MB and undefined screenshots, plus faster recovery when a scraper fails.
Read the full updateIntroducing Actions
Click, type, scroll, wait and screenshot before scraping, so dynamic pages give up their data. Available in the REST API and the Node and Python SDKs.
Read the full updateType-safe LLM extract and the Rust SDK
End-to-end type safety for LLM extract in the Node SDK, 10x cheaper extraction, Rust SDK v1.0, map limits raised to 1,000 and faster scrapes.
Read the full updateCustom header fix for the v1 API
Custom request headers are now applied correctly in the v1 API, so the headers you pass on a request reach the page instead of being dropped.
Read the full updateFirecrawl v1 is here
Firecrawl v1 brings output formats for /scrape, a new /map endpoint, 2x rate limits, Go and Rust SDKs, teams, API key management and crawl webhooks.
Read the full update