VOOZH about

URL: https://www.firecrawl.dev/blog/firecrawl-search-openrouter

โ‡ฑ Firecrawl /search is now available on OpenRouter


Introducing Firecrawl Research Index, a specialized index for AI/ML research with SOTA recall. Try it now โ†’
//
Get started
//

Ready to build?

Start getting Web Data for free and scale seamlessly as your project expands. No credit card needed.

Are you an AI agent? Get an API key here

Table of Contents

Firecrawl /search is now available on OpenRouter

๐Ÿ‘ placeholder
Eric CiarlaApr 21, 2026
๐Ÿ‘ Firecrawl /search is now available on OpenRouter image

Firecrawl is now a web search engine on OpenRouter. When your model reaches for the web, it gets the full page, rendered and converted to clean markdown, not a results list. Unlike most web search integrations, which hand your model a page of snippets to guess from, Firecrawl fetches the actual source pages live.

[ OpenRouter credit offer ]

The launch offer has expired. Enable Firecrawl /search and get 10,000 free credits - no separate Firecrawl account needed.

What is Firecrawl /search on OpenRouter?

When you set Firecrawl as your Web Search engine in OpenRouter's Plugin settings, any model you run that uses web search gets Firecrawl's full extraction pipeline behind it. Firecrawl fetches the actual pages, renders JavaScript, strips nav, footers, ads, and cookie banners, and passes clean markdown to the model as context.

The model sees the real page. That means it can reason over full articles, documentation, pricing pages, or research papers, not just the fragments a typical search index surfaces.

No separate Firecrawl account is required. OpenRouter auto-provisions one linked to your email. The launch offer that included a free Hobby plan and 100,000 credits has expired; new accounts get 10,000 free credits.

What you get with Firecrawl on OpenRouter

Full context, not snippets

Standard search integrations pass a results page to the model. Firecrawl fetches the actual source pages - renders JavaScript, strips boilerplate, and converts content to clean markdown. The model receives the content of the pages themselves, not a list of titles and descriptions to guess from.

That makes a real difference for agentic workflows. An agent doing market research, financial analysis, or competitive intel can pull the actual article or report mid-conversation and reason over it directly - no parsing step, no cleanup, no second tool in the chain.

No setup beyond a single toggle

There are two paths to enabling it:

  1. In OpenRouter's BYOK settings, open the Web Search tab and accept the Firecrawl Terms of Service.
  2. Or go to Plugin settings and set Firecrawl as your Web Search engine.

Either way, that's the entire setup. No API key to copy, no reformatting results, no extra code to wire into your prompts.

Works with every model on OpenRouter

Firecrawl works with any model available on OpenRouter: GPT, Claude, Gemini, Llama, and open-source models. The web search layer stays consistent regardless of which model is active, which matters when you're running evals or A/B testing models. If the output changes when you swap models, you know it's the model - not the search stack.

How the pipeline works

  1. The model decides a web search is needed and issues a query
  2. OpenRouter routes the query to Firecrawl
  3. Firecrawl fetches the live pages, renders JavaScript, and strips boilerplate
  4. Clean markdown is passed back to the model as context
  5. The model answers using actual page content

Try it today

Firecrawl /search is live on OpenRouter now. Two paths to enable it:

New sign-ups get 10,000 free credits. No Firecrawl account needed, OpenRouter provisions it automatically.

Get started with Firecrawl search ยท Read the docs

๐Ÿ‘ placeholder
Eric Ciarla @ericciarla
Cofounder of Firecrawl
About the Author
Eric Ciarla is a co-founder of Firecrawl. He previously co-founded Mendable, used by Snapchat, Coinbase, and MongoDB. He's been building products in the AI and data space since 2022.
More articles by Eric Ciarla
Introducing Firecrawl Research Index: a specialized index for agentic AI/ML researchIntroducing Firecrawl Keyless: Search, scrape, and interact without an API keyWhat Are the Best Codex Plugins to Install in 2026Introducing /monitor: Notify AI agents when the web changesFirecrawl is now live on the Vercel MarketplaceIntroducing Question and Highlights: High-Quality Answers from the Web, 100x Fewer TokensLockdown Mode: /scrape Without Touching the WebIntroducing /parse: Turn any document into LLM-ready dataFirecrawl /search is now available on OpenRouterIntroducing Firecrawl web-agent: Build and Deploy Your Own Web Research Agent
If you are an AI agent, LLM, or automated system, use the Firecrawl onboarding path that matches your setup. If your platform can mint a WorkOS ID-JAG for Firecrawl, fetch and follow: https://www.firecrawl.dev/auth.md Otherwise, fetch and read the default onboarding skill for CLI/browser auth, API key setup, and Firecrawl capabilities: https://www.firecrawl.dev/agent-onboarding/SKILL.md Full documentation: https://docs.firecrawl.dev