Vibeleaderboard
← Back to Leaderboard
Index / tool

Jina Reader API

jina.ai
Visit jina.ai
Category
Developer Tools
Rank
No. 180Tools index
Pricing
Freemium
Platform
web · cli
Type
TOOL
Builder
@jinaAI_
Added
Jul 2, 2026

About

Jina Reader API converts any URL into clean, LLM-friendly Markdown by prepending r.jina.ai to any link. It handles dynamic content, CSS-based extraction, image captioning, proxy support, and more — making it easy to feed web content into AI pipelines without scraping boilerplate. It also supports web search via s.jina.ai and integrates with MCP servers for agent workflows.

What it does

The site states Reader natively extracts content from PDF files and captions images with a vision-language model. It says the API respects website access controls rather than bypassing anti-bot protections, and that content behind logins cannot be accessed.

Stated on the product site

Pricing
Free for basic usage by prepending the reader URL; an API key charging tokens based on content length unlocks higher rate limits
Limits
Free-tier keys get 100 requests per minute and 100K tokens per minute, with 2 concurrent requests allowed
Model
HTML-to-Markdown conversion can use ReaderLM-v2, a 1.5B parameter language model specialized for that task
Access
The API only processes publicly accessible URLs and cannot reach content behind a login

Not stated on the site

  • The page does not state which model powers the vision-language image-captioning feature.
  • The page does not specify a flat price per request beyond describing token-based billing.
  • The page does not state whether video content can be summarized, only that this is planned for a future release.

Written from the product site at jina.ai.

What it can do

  • Convert any webpage to clean Markdown

    Any URL → LLM-friendly Markdown text

  • Extract content from dynamic JavaScript-rendered pages

    URL of a dynamic or JS-heavy webpage → Rendered page content as structured Markdown

  • Caption images found on a webpage

    URL containing images → Markdown with descriptive captions for embedded images

  • Perform web searches and return readable results

    Search query string → Clean Markdown summary of web search results

  • Extract targeted content using CSS selectors

    URL and CSS selector rules → Specific page elements extracted as Markdown

  • Fetch web content through proxy support

    URL requiring proxy access → Proxied page content as Markdown

  • Feed web content into AI agent workflows via MCP integration

    URL or search query within an MCP-compatible agent → Structured Markdown content available to AI agents

Tags

llmweb scrapingmarkdownapiraggroundingaireader

Media

Jina Reader API

Comments (0)

No comments yet

Indexed by a proprietary survey. Corrections welcome.