Jina Reader turns a URL into LLM-friendly Markdown
when you put https://r.jina.ai/ in front of it. Readmint does the same job
with a JSON API. Both strip navigation, headers, footers and ads, both load
pages built by JavaScript in a headless browser, and both refuse to get past
anti-bot protection. They differ in the extras. Jina Reader's features are as
its page described them on 31 October 2026.
At a glance
| Readmint | Jina Reader | |
|---|---|---|
| How you call it | POST /v1/read with JSON |
Prefix the URL with r.jina.ai; options as headers |
| Use without an account | No: free plan with a key | Yes, at a lower rate limit |
| JavaScript pages | Fetched directly first, a browser only when needed | A headless browser by default; a direct fetch on request |
| One part of a page by CSS selector | Yes | Yes, and removing parts too |
| PDFs | No | Yes |
| Image captions from a vision model | No | Yes |
| Web search | No | Yes (s.jina.ai) |
| Link previews with image size and oEmbed | Yes | No dedicated endpoint |
| A site's URLs from its sitemaps | Yes (map) |
No dedicated endpoint |
| MCP server | Yes | Yes |
| Pricing | Pages a month on a plan; failed pages free | Tokens; failed requests free |
Choose Jina Reader if
- You want to try a URL in a browser or with curl without signing up.
- You need PDFs, image captions or web search.
- You'd rather pay by tokens than by pages.
Choose Readmint if
- You want plain per-page pricing: one page is one credit, however long it is and whether or not it needed the browser.
- You also need link previews (with image sizes and oEmbed) or a list of a site's pages from its sitemaps.
- You want fast reads of ordinary pages: Readmint fetches the HTML directly first and starts a browser only when the page needs one.
Try it
The free plan includes 1,000 pages a month, no card needed. See how reading works for every option.