Firecrawl raises $75M Series B and launches Alexandria knowledge library for AI agents
Tags VC · Startups · AI / ML
Firecrawl, a web-data platform for AI agents, raised $75 million in Series B funding and launched Alexandria, a unified search interface that consolidates live web search, official data providers, site-specific connectors and Firecrawl's own indexes. The company said agents using Alexandria scored 21% higher on answer quality than those using built-in web tools, based on blind AI judging across 845 tasks with the same model and prompts. Alexandria provides access to more than 100 data providers, with discovery and inspection free and execution billed at each tool's listed price. It is available in Claude, Claude Code, ChatGPT and Codex through Firecrawl plugins, as well as via API, MCP and CLI.
Technical significance
Alexandria bundles retrieval, connectors and paid data providers behind one interface, which could simplify how agent builders wire up external knowledge and reduce the token cost of extracting specific data from pages. The reported 21% answer-quality gain across 845 tasks is a vendor-run evaluation, so it should be read as a directional claim rather than an independent benchmark. The $75M raise, alongside the earlier Research and Developer indexes, positions Firecrawl as infrastructure for the retrieval layer of agentic applications.