R Jina

Reader is a web-content extraction API built by Jina AI. It converts any public URL into clean, LLM-ready Markdown by prefixing the target address with r.jina.ai/ — either directly in a browser address bar or via an API request. Under the hood it fetches the page, strips navigation, ads, headers, and footers using Mozilla's Readability library, and reformats the remaining main content as Markdown for use as grounding context for language models, RAG pipelines, and agents. Beyond basic web pages, Reader also handles PDFs, Office documents, and images (using AI captioning to describe visual content), and a companion endpoint, s.jina.ai, performs web search and returns Markdown-formatted results. Request headers let developers target content with CSS selectors, set timeouts, control token budgets, and choose response formats such as plain text, HTML, screenshots, or structured JSON. Reader is aimed at developers and AI engineers who need reliable, structured web content for feeding into large language models, rather than general consumers looking for a chat interface. Free, unauthenticated use is available but rate-limited (around 20 requests per minute); registering for a free API key raises the limit substantially, and paid API keys unlock higher throughput and token allowances.