Firecrawl

Firecrawl is a web crawling, scraping, and search API for developers and AI teams that need websites converted into LLM-ready data for agents, RAG, and data pipelines.

Key capabilities include:

• Scrape and crawl at scale, including JavaScript-heavy and dynamic pages, typically billed at 1 credit per page
• Clean output formats for downstream AI use, including Markdown and JSON (plus structured data workflows)
• AI-powered extraction via the /extract endpoint for turning pages into structured fields
• Agent-based web data gathering for prompt-driven collection tasks (with dynamic credit usage)

Pricing: Freemium with credit-based plans, Free (500 one-time credits), Hobby ($16/month, 3,000 credits), Standard ($83/month, 100,000 credits), Growth ($333/month, 500,000 credits), Scale ($599/month, 1M credits), Enterprise (custom). Separate token-based pricing applies to /extract (from $89/month for 18M tokens/year).

What makes it notable: It is built for AI workflows with LLM-friendly outputs, built-in proxy management and ban prevention, integrations with tools like LangChain and LlamaIndex, and an open-source option for self-hosting alongside the hosted API.