Millions of requests for Wikipedia data flow through Firecrawl every month. It's one of the most requested sources on our platform, so we partnered with Wikimedia Enterprise.

Starting today, Firecrawl routes all Wikipedia requests through the Wikimedia Enterprise On-demand API, a paid, commercial API designed for high-volume programmatic access. We're paying for this access directly, contributing financially to the infrastructure that keeps Wikipedia running. This is how Firecrawl approaches the web: smart caching to keep traffic efficient, clean data so models use less energy and fewer tokens, and now direct partnerships that compensate the people behind the content.

On your end, requests are faster because there's no headless browser spinning up to render pages. Data is more consistent because the Enterprise API returns structured, clean content instead of raw HTML. All supported languages are covered through a single endpoint. Nothing changes in how you call the API.

const result = await app.scrape("https://en.wikipedia.org/wiki/NASA", {
  formats: ["markdown"],
});

Source URLs come back with every response, so attribution is built in. Wikipedia is the closest thing to a ground truth layer the open web has, and this partnership makes it the most reliable source on our platform too.

This is the first data partnership like this. Wikimedia Enterprise was the obvious place to start because of the volume, but the model is the same everywhere: efficient access, fair compensation, and less unnecessary load on the sites we depend on. Between partnerships like this one and a Creator Program we're developing to compensate content creators directly, the goal is straightforward: the people producing content on the open web should benefit from the traffic, not just absorb it. Agents accessing the web are growing at an incredible rate, and Firecrawl is at the heart of it, enabling agents and the humans behind them to get the data they need. That means the infrastructure connecting them to publishers needs to grow with it, and we want Firecrawl to be the standard for how that works.

Ready to get started with Wikipedia data? Try Firecrawl today.