Source: Jina Reader (r.jina.ai / s.jina.ai)
Jina AI's Reader API that converts any URL into LLM-friendly markdown by URL prefix, plus a search endpoint, control headers, ReaderLM-v2 and token-based pricing.
Contents
- What: product and API documentation page
- Author / publisher: Jina AI
- Date: read 2026-11-14 (living page)
- Original: jina.ai/reader
Summary
Prefixing a URL with r.jina.ai/ returns the page’s main content as markdown with navigation, ads and footers stripped; s.jina.ai runs a search and returns the top five results with cleaned content.
Key claims
- Headers control behaviour:
X-Engine(directplain fetch fastest; default renders JavaScript in a headless browser),X-Target-Selector(CSS selectors),X-With-Generated-Alt(captions images),X-No-CacheorX-Cache-Tolerance,X-Timeout,X-Respond-Withfor output shape. - ReaderLM-v2: a 1.54B-parameter model converting HTML to markdown or JSON, selectable per request, with schema or instruction headers for structured extraction.
- PDFs are handled natively, with page targeting.
- Rate limits: no key 20 RPM for Reader and search blocked; free/paid key 500 RPM Reader, 100 RPM search; premium 5,000 / 1,000 RPM.
- Billing is by output tokens; each search request costs a fixed amount starting at 10,000 tokens; failed requests are free; new keys get 10M free tokens.