Jina AI Reader: vendor profile

In short

Reader works from the caller's side and adds no llms.txt or markdown route to the site it reads. It allows 20 requests a minute without an API key and 500 with a free or paid key. Elastic has sold the commercial model licence as its own SKU since 2026-08-10, and procurement for the hosted APIs runs through Elastic.

At a glance

standards.txt jina-reader

llms-txt               = no statement
markdown-negotiation   = no statement
markdown-twin          = no statement
initial-html           = no statement
real-404               = no statement
vary-accept            = no statement

0 supports, 0 partial or announced, 6 no statement.

Jina AI Reader Markdown or llms.txt tool

Hosted API that fetches any URL in a browser and returns LLM-ready markdown or JSON when the URL is prefixed with r.jina.ai, plus a search endpoint at s.jina.ai; it shows a site what an agent pipeline extracts from its pages.

Type: Markdown or llms.txt tool
Operator, as the facts below state it: Jina AI
Home page: jina.ai/reader, answered HTTP 200 on 2026-11-05
Last read: 5 November 2026

What it does for this layer

Jina AI Reader converts a page to LLM-friendly input when r.jina.ai is prepended to its URL. [primary source] source, read 5 November 2026.
Jina's s.jina.ai endpoint searches the web and returns the top five results with their URLs and contents as LLM-friendly text. [primary source] source, read 5 November 2026.
The Reader API rate limit table lists r.jina.ai at 20 requests per minute without an API key, 500 with a free or paid key and 5,000 with a premium key, with an average latency of 7.9 seconds. [primary source] source, read 5 November 2026.
Reader lets the caller choose the browser engine used to fetch a page, override the User-Agent string, set the browser locale and set the window size. [primary source] source, read 5 November 2026.
Reader's markdown output options for headings, rules, emphasis and links are passed through to the Turndown library. [primary source] source, read 5 November 2026.
Reader caches fetched URL contents for a period; callers can bypass the cache, accept cached content younger than N seconds, or ask that a request be neither cached nor logged. [primary source] source, read 5 November 2026.
Jina says Reader reads PDFs natively, and an OCR option using jina-ocr-v1 turns images into markdown at 40 times the token cost. [primary source] source, read 5 November 2026.
Jina says the Reader service code is on its GitHub organization while the ReaderLM-v2 and jina-vlm models are CC-BY-NC 4.0, and that Elastic has sold the commercial licence as its own SKU since 2026-08-10. [primary source, 2026-08-10] source, read 5 November 2026.
Jina's FAQ routes purchase orders, negotiated contracts and procurement for the hosted APIs through Elastic rather than self-serve Stripe billing. [primary source] source, read 5 November 2026.

Products and plans

Plans and prices as the vendor's own pages showed them on the date read.
Product	What it is	Price	As of	Source
Reader API (r.jina.ai)	URL to markdown, text, HTML or JSON conversion with browser rendering, caching and PDF support.	Usage-based, with a free allowance Usable without a key at 20 requests a minute; paid token packs raise limits	2026-11-05	source
Search API (s.jina.ai)	Web search that returns the top five results with their page contents as text.	No price on the pages read

Standards

The six standards this site tracks; no statement means the pages read said nothing.
Standard	Status	What the vendor's pages say	Source
llms-txt	no statement	Reader does not generate llms.txt for sites in the page read.	none
markdown-negotiation	no statement	Reader converts pages itself rather than negotiating with the origin, in the page read.	none
markdown-twin	no statement	Not a site feature; Reader produces its own markdown copy on request.	none
initial-html	no statement	Reader runs a browser, so it reads JavaScript-rendered content that a plain fetch would miss.	none
real-404	no statement	Not covered in the page read.	none
vary-accept	no statement	Not covered in the page read.	none

Checks on its own site

Plain-fetch checks ran on https://jina.ai/reader/; the Ora and Cloudflare scores cover the whole host jina.ai.

Readings of jina.ai, the vendor's own website, which describe that site and not the product.
Reading	Result	Date
Home page request	HTTP 200; H1 in the initial HTML: yes; 5,153 words; 0 JSON-LD blocks	2026-11-05
Request with Accept: text/markdown	HTTP 200, text/html, Vary: Accept-Encoding	2026-11-05
/llms.txt at the host root	HTTP 404, text/html, 29 KB	2026-11-05
A made-up path	HTTP 404	2026-11-05
Ora score, cached	61, grade C	2026-10-01
Cloudflare keyless scan	20, L1 Basic Web Presence	2026-11-05
Shared crawl row	outside the ranked slice; home page HTTP 200; H1 in the HTML: yes; 1,596 words; JSON-LD types: none; answers markdown: no; made-up path HTTP 404; llms.txt valid: no	2026-11-04

What it does not do

Jina AI Reader changes nothing on the site it reads; it is a client-side conversion service, so a site gains no llms.txt or markdown route from it.
Reader's commercial model licence is not included in token purchases, according to Jina's FAQ.

Related pages on this site

Sources

Reader API - Jina AI https://jina.ai/reader/, read 5 November 2026

Last reviewed 5 November 2026