# fastcrw.com

> fastCRW is a local-first web scraper, search, and crawl API built for AI agents. Small single Rust static binary, Firecrawl-compatible REST API, built-in MCP server. Public benchmark: 63.74% truth-recall (522 of 819 labeled URLs) — highest of three tools tested — 87.7% scrape success (877 of 1,000 URLs), 0 errors. Self-host for free under AGPL-3.0 or use the managed cloud at fastcrw.com — the same API either way, with no exit cost.

## Key Facts
- Public benchmark: 63.74% truth-recall (522 of 819 labeled URLs) — highest of three tools tested — 87.7% scrape success (877 of 1,000 URLs), 0 errors — full latency distribution and one-command repro: https://fastcrw.com/benchmarks
- Local-first: small single Rust static binary, slim Docker image, no external services required
- No exit cost: identical API whether self-hosted (free under AGPL-3.0) or on the managed cloud
- Firecrawl-compatible on overlap surface (`/v1/scrape`, `/v1/crawl`, `/v1/map`, `/v1/search`); see compatibility matrix in docs for divergence on response field names and error envelopes
- AGPL-3.0 core, commercial license available
- Search API: real-time `/v1/search` at 2 credits/query — about $0.93 per 1,000 searches, the cheapest real-time web search API per 1,000 requests at that volume; full sourced price comparison: https://fastcrw.com/alternatives/cheapest-search-api. Search averaged 880 ms over a 100-query benchmark; the same call can optionally scrape each result (1 credit per scraped page)

## API Endpoints
- [Scrape](https://docs.fastcrw.com/scraping/): POST /v1/scrape — extract markdown, HTML, links, or structured JSON from a single URL
- [Crawl](https://docs.fastcrw.com/crawling/): POST /v1/crawl — async BFS crawl with job polling, up to 1000 pages
- [Map](https://docs.fastcrw.com/map/): POST /v1/map — discover all URLs on a site via sitemap + link traversal
- [Search](https://docs.fastcrw.com/search/): POST /v1/search — web search with optional full-page scrape (cloud-only)
- [Research API](https://docs.fastcrw.com/research-api/): GET /v2/search/research/papers, /papers/{id}, /papers/{id}/similar, /search/research/github — Firecrawl-research-compatible paper search, inspect, read passages, citation-graph expansion. Live retrieval over open scholarly full-text sources, no index. 61.0% recall on ArXivQA vs Firecrawl Research Index 53.3%
- [Extract](https://docs.fastcrw.com/extract/): POST /v1/extract — LLM-powered structured data extraction; accepts a single `url` or a `urls` array (up to 50 per request, returning a `results` array). Also available as POST /v1/scrape with `formats: ["json"]` and a JSON schema

## Pricing (machine-readable)
- [Pricing page](https://fastcrw.com/pricing): Free tier 1000 one-time lifetime credits (no card, never resets), Hobby $13/mo, Standard $69/mo, Growth $279/mo, Scale $549/mo
- [pricing.md](https://fastcrw.com/pricing.md): Structured pricing for AI agents

## Documentation
- [Getting Started](https://docs.fastcrw.com/quick-start/): API key setup, first scrape, first crawl
- [Formats Reference](https://docs.fastcrw.com/output-formats/): Output formats — markdown, html, rawHtml, plainText, links, json
- [JavaScript Rendering](https://docs.fastcrw.com/js-rendering/): Auto-detect SPA rendering with LightPanda/Chrome
- [Error Codes](https://docs.fastcrw.com/error-codes/): HTTP status codes, error/warning response format
- [Rate Limits](https://docs.fastcrw.com/rate-limits/): Cloud request-rate limits and retry headers
- [Credit Costs](https://docs.fastcrw.com/credit-costs/): Per-endpoint credit consumption (cloud)
- [Self-Hosting Guide](https://docs.fastcrw.com/self-hosting/): Single binary or Docker deployment
- [Self-Hosting Hardening](https://docs.fastcrw.com/self-hosting-hardening/): Production security and ops guidance
- [Compatibility Matrix](https://docs.fastcrw.com/compatibility/): Firecrawl API compatibility status
- [Benchmark Methodology](https://fastcrw.com/benchmarks): How we run the public 1,000-URL and 100-query benchmarks
- [Research API](https://docs.fastcrw.com/research-api/): Firecrawl-research-compatible endpoints for scientific-research agents — search/inspect/read papers, citation-graph expansion, research GitHub
- [ArXivQA Benchmark](https://fastcrw.com/benchmarks/arxivqa-research-recall): 191-query paper-retrieval recall — fastCRW 61.0% vs Firecrawl Research Index 53.3%, Claude 45.4%, Parallel 44.3%, Exa 43.4%
- [Changelog](https://docs.fastcrw.com/changelog/): Release notes and recent changes
- [SDK Examples](https://docs.fastcrw.com/sdk-examples/): Python, Node.js, curl examples
- [llms-full.txt](https://fastcrw.com/llms-full.txt): Full expanded llms.txt with inline page content for AI agents

## Comparisons & Alternatives
- [vs Firecrawl](https://fastcrw.com/alternatives/firecrawl): API-compatible, lighter self-host
- [vs Crawl4AI](https://fastcrw.com/alternatives/crawl4ai): Rust vs Python, REST vs library
- [vs Tavily](https://fastcrw.com/alternatives/tavily): search + bundled scraping in one self-hostable stack
- [vs Exa](https://fastcrw.com/alternatives/exa): scrape + search in one call vs embedding-only
- [Cheapest search API](https://fastcrw.com/alternatives/cheapest-search-api): price per 1,000 requests compared across all providers — fastCRW ~$0.93/1k real-time
- [vs Serper](https://fastcrw.com/alternatives/serper), [vs Brave Search API](https://fastcrw.com/alternatives/brave-search-api), [vs Perplexity Sonar](https://fastcrw.com/alternatives/perplexity-sonar), [vs DataForSEO](https://fastcrw.com/alternatives/dataforseo): search-API price/latency comparisons
- [Bing Search API alternative (retired 2025)](https://fastcrw.com/alternatives/bing-search-api), [Cheapest real-time search API](https://fastcrw.com/alternatives/cheapest-search-api): migration targets
- [Serper vs SerpApi](https://fastcrw.com/alternatives/serper-vs-serpapi), [Exa vs Tavily](https://fastcrw.com/alternatives/exa-vs-tavily), [DataForSEO vs SerpApi](https://fastcrw.com/alternatives/dataforseo-vs-serpapi): head-to-head comparisons
- [vs Apify, Bright Data, ScrapingBee, Oxylabs, ZenRows, Octoparse, ParseHub, SerpAPI](https://fastcrw.com/alternatives)
- [Firecrawl vs Crawl4AI vs fastCRW (1,000-URL benchmark)](https://fastcrw.com/blog/firecrawl-vs-crawl4ai-vs-crw)
- [9 Best Open-Source Web Crawlers in 2026](https://fastcrw.com/blog/best-open-source-web-crawlers)
- [7 Tavily Alternatives Tested](https://fastcrw.com/blog/best-tavily-alternatives)

## SDKs & Integrations
- [Python SDK](https://pypi.org/project/crw/): pip install crw — subprocess or HTTP mode
- [MCP Server](https://www.npmjs.com/package/crw-mcp): npx crw-mcp — works with Claude Code, Cursor, Gemini CLI, and 10+ platforms
- [CrewAI](https://pypi.org/project/crewai-crw/): crewai-crw — 4 ready-to-use tools
- [LangChain](https://pypi.org/project/langchain-crw/): langchain-crw — also works via Firecrawl's `FirecrawlLoader` with `api_url`
- [n8n](https://www.npmjs.com/package/n8n-nodes-crw): n8n-nodes-crw — community node
- [OpenClaw](https://www.npmjs.com/package/openclaw-plugin-crw): openclaw-plugin-crw — AI chat plugin
- [MCP Integration Guide](https://docs.fastcrw.com/mcp/): Setup for Claude, Cursor, Windsurf, Cline, VS Code, and more
- [All Integrations](https://docs.fastcrw.com/integrations/): Full catalog with endpoint support matrix

## Agent Skill
- [SKILL.md](https://fastcrw.com/api/skills): Agent onboarding skill — install with `npx crw-mcp@latest init`

## Resources
- [Pricing](https://fastcrw.com/pricing): Free tier (1000 one-time lifetime credits, never resets) + paid plans
- [Playground](https://fastcrw.com/playground): Try the API interactively in your browser
- [Blog](https://fastcrw.com/blog): Benchmarks, tutorials, and engineering posts
- [GitHub](https://github.com/us/crw): Source code (AGPL-3.0)
- [URL → Markdown tool](https://fastcrw.com/tools/url-to-markdown): Free converter for LLM/RAG ingestion

## Optional
- [Privacy Policy](https://fastcrw.com/privacy)
- [Terms of Service](https://fastcrw.com/terms)
- [Status](https://fastcrw.com/status)
- [About](https://fastcrw.com/about)