Hiba Fathima
Firecrawl turns any website into clean, LLM-ready data. Eden AI is a unified API to 500+ AI models. Together, you get the whole pipeline behind one key: Firecrawl handles the web data, Eden AI handles the models, billing, and everything else in your stack.
Today, we're announcing that Firecrawl is available directly inside Eden AI as a first-class web data provider. No separate sign-up, no second API key to track down.
Add Firecrawl to your AI stack with one key. No separate Firecrawl account needed to get started.
Clean web data, alongside the models you already call
If you're building RAG pipelines, agents, or research tools, you need two things: fresh data from the web and a model to reason over it. That usually means two vendors, two dashboards, and glue code in between.
Firecrawl was built for the first half. It converts any webpage into structured, LLM-ready output, including JavaScript-rendered and dynamic pages, so you can pass the result straight to a model without wasting tokens on noisy HTML.
Eden AI covers the second half, with Claude, GPT, Gemini, Mistral, and 500+ other models behind a single API and a single bill. Firecrawl now sits in that same catalog under the web feature, with seven capabilities live:
- Web Scraping (
web/scraping/firecrawl) - Fetch a single URL and get clean markdown, text, or HTML - Web Search (
web/search/firecrawl) - Search the web and get ranked results with title, URL, and content - Site Map (
web/map/firecrawl) - Discover the list of URLs on a site - Web Crawl (
web/crawl_async/firecrawl) - Crawl a site from a starting URL and get the content of its pages - Batch Web Scraping (
web/batch_scrape_async/firecrawl) - Scrape many URLs in one request - Structured Extraction (
web/structured_extraction_async/firecrawl) - Turn pages into JSON against a prompt, a schema, or both - Deep Research (
web/research_async/firecrawl) - Run multi-step web research on a query and get back findings with sources
How it works
Every call goes through the standard Eden AI endpoint with one EDEN_AI_API_KEY, using the same feature/subfeature/provider pattern as the rest of the platform.
Scraping, search, and map are synchronous. Post to /v3/universal-ai and the content comes back in the response:
curl -X POST https://api.edenai.run/v3/universal-ai \
-H "Authorization: Bearer $EDEN_AI_API_KEY" \
-H "Content-Type: application/json" \
-d '{ "model": "web/scraping/firecrawl", "input": { "url": "https://firecrawl.dev" } }'Crawl, batch scrape, structured extraction, and deep research run as jobs. Post to /v3/universal-ai/async, then poll /v3/universal-ai/async/{public_id} until the status flips to success.
Responses use the same envelope as every other Eden AI feature, including a cost field per call, so your web data and your model calls show up on one bill.
You can browse the entire catalog, including every Firecrawl model string, its mode, and its pricing, before signing up for anything:
curl https://api.edenai.run/v3/info/webWhat you can build with Firecrawl + Eden AI
RAG on live web data. Crawl or scrape a set of sources with Firecrawl, embed the markdown, and answer questions with any model in the catalog. One key covers the whole loop.
Web-augmented agents. Give an agent search and scraping as tools next to its LLM, so it can go find what it doesn't know instead of guessing.
Structured extraction at scale. Batch scrape a list of URLs, run structured extraction with a prompt or schema, and post-process the JSON with a model. Useful for lead enrichment, competitive tracking, and catalog monitoring.
Market monitoring. Map and crawl a domain on a schedule, then have a model summarize what changed since the last run.
Autonomous research. Hand deep research a plain-language brief and it searches and reads the web on its own, returning what it found with sources, shaped to a schema if you pass one.
Ready to add live web data to your AI stack?
Get started on the Firecrawl provider page on Eden AI, or explore what each endpoint returns in the Firecrawl docs.
Frequently Asked Questions
Do I need a Firecrawl account to use Firecrawl through Eden AI?
No. Calls go through the standard Eden AI endpoint with a single EDEN_AI_API_KEY, and usage is billed on your Eden AI account. You only need a separate Firecrawl key if you want to call Firecrawl directly.
Which Firecrawl capabilities are available in Eden AI?
Seven: web scraping, web search, site map, web crawl, batch web scraping, structured extraction, and deep research. Scraping, search, and map run synchronously. Crawl, batch scrape, structured extraction, and deep research run as async jobs that you submit and then poll.
How do I see the exact model strings and pricing?
Eden AI exposes a public discovery endpoint at api.edenai.run/v3/info/web that returns every subfeature, its display name and description, whether it is sync or async, the exact model string to paste, and per-model pricing and regions. No account or key is required to read it.
Is this the same as Eden AI's LLM web search option?
No. Eden AI's web_search_options flag turns on a model's own built-in search during an LLM call. The Firecrawl integration is a separate web feature that returns the actual page content, so you control what gets scraped, crawled, and passed to the model.
Where do I track usage and costs?
In your Eden AI dashboard. Every response carries a cost field with the credits charged for that call, using the same envelope as every other Eden AI feature, so web data and model calls land on one bill.
Where does Eden AI run, and what does that mean for GDPR?
Eden AI is a French company and its gateway runs on EU infrastructure with GDPR-aligned processing under a single contract covering every provider behind it, which is one less vendor review for European teams. Each provider lists its own region in the public discovery endpoint, so you can check where a given model runs before you call it.
