# Retriever AI services

> Retriever AI runs a cloud browser cluster that loads web pages and returns what an agent needs to read them: the visible text, the accessibility tree, and a map of every link on the page. Pages load in real browsers behind rotating proxies, so sites that rate limit or block plain HTTP fetches still come back, and no language model runs on the way out, so the output is raw page data ready for your own extraction, search, or retrieval pipeline. Each call takes one or more URLs and returns one tab per URL in the same order. Every call is one request at a flat rate.

1 service. Open its page for the full price matrix, plans, how to call it, and the API reference.

## Scrape

Loads one or more URLs in Retriever AI's cloud browser cluster and returns, per URL, the visible text, an accessibility tree with a numeric id on every link, and a map from those ids to the link targets. Real browsers behind rotating proxies, no model in the loop. Each call is one request.

- From $0.05 / request
- Billed by: request
- Plans: Pay per request (pay as you go)
- Service page: https://agents.rtrvr.ai/zcj/q5wlfmj8vety/services/scrape
- Service spec: https://agents.rtrvr.ai/zcj/q5wlfmj8vety/services/scrape/openapi.json