Link Preview API: Open Graph Title, Image & Description
Pricing
from $0.91 / 1,000 link previews
Link Preview API: Open Graph Title, Image & Description
Link preview API for up to 1,000 URLs per run: Open Graph and Twitter Card title, description, absolute image URL, site name, favicon, canonical URL and oEmbed link, one row per URL. Plain HTTP, no browser, fast. Charged only for URLs with a preview; errors are free. Empty input = free demo.
Pricing
from $0.91 / 1,000 link previews
Rating
0.0
(0)
Developer
Actor Factory
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
a day ago
Last modified
Categories
Features
Share
Link Preview API: send a list of URLs and get one link-preview row per URL — Open Graph and Twitter Card title, description, absolute image URL, site name, favicon, canonical URL and oEmbed link. Plain HTTP, no browser, built for high-volume link unfurling.
| inputUrl | title | description | image | siteName |
|---|---|---|---|---|
| https://apify.com | Apify: Marketplace of ready-to-run tools for AI | Thousands of tools to automate your business. Get real-time web data, … | https://apify.com/og-image?title=… | Apify |
(real output for apify.com on 2026-11-12, shortened to 5 of the columns; the full row also has finalUrl, statusCode, imageWidth/Height (1200 × 630), type, url, favicon, twitterCard, oEmbedUrl, lang and error)
For developers who show link previews — chat apps, CMSs, newsletters, bookmark managers, social schedulers, AI agents — and need the title, description and image of many URLs through one API call, without running a headless browser.
Try it: press Start with the prefilled example (3 URLs) and you get 3 rows in a few seconds, for less than a cent. Clear the URL list and press Start for a free demo on 3 sample URLs (rows carry _demo: true, not charged).
What you get
- One row per URL, in input order:
inputUrl,finalUrl,statusCode,title,description,image,imageWidth,imageHeight,siteName,type,url(canonical),favicon,twitterCard,oEmbedUrl,lang,error. - Sensible fallbacks so you always get the best preview the page offers:
- title:
og:title→twitter:title→<title>; - description:
og:description→twitter:description→<meta name="description">; - image:
og:image→og:image:secure_url→twitter:image→<link rel="image_src">; - canonical:
<link rel="canonical">→og:url; site name:og:site_name→application-name; - favicon:
<link rel="icon">→apple-touch-icon→ the site's/favicon.ico(not checked); - language:
<html lang>→Content-Language→og:locale.
- title:
- Absolute URLs: relative, protocol-relative (
//cdn…) and<base href>image, favicon, canonical and oEmbed URLs are resolved against the final URL.data:andjavascript:URLs are ignored. - oEmbed discovery: the oEmbed endpoint the page advertises (
application/json+oembed, else XML), ready to call for rich embeds. - Any language: the charset is taken from the HTTP header, a byte-order mark or
<meta charset>, so Japanese, Cyrillic, Greek or Chinese pages come out right; undeclared legacy pages are read as Windows-1252. - Fast and lean: plain HTTP (no browser), up to 50 URLs in parallel, keep-alive connections, and at most 1 MB of HTML read per page — reading stops early once the
<head>has everything. - Safe by design: URLs that resolve to internal addresses (localhost, private ranges, link-local such as 169.254.169.254, IPv6 loopback/unique-local) are never contacted — also when a public URL redirects to one. These rows get
error: "blocked_private_address". - Never fails the whole run on bad input: DNS failures, timeouts, 404s, non-HTML files and invalid URLs each become a row with an
errorcode — and are not charged.
How to use
- Put your URLs in URLs, one per line (up to 1,000 per run).
- Optional: Timeout (default 15 s), Concurrency (default 10), User-Agent.
- Press Start, or call the Actor from your code. For link unfurling in an app, the synchronous endpoint returns the rows directly:
curl -X POST "https://api.apify.com/v2/acts/ucchiii~link-preview-api/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \-H "Content-Type: application/json" \-d '{"urls": ["https://apify.com", "https://www.python.org"]}'
{"urls": ["https://apify.com", "https://www.python.org", "https://github.com/apify/crawlee"],"timeoutSecs": 15,"concurrency": 10}
Output fields
inputUrl— the URL as given;finalUrl— the page that was read after redirects (up to 10);statusCode— HTTP status of that page.title,description,image,imageWidth,imageHeight,siteName,type(og:type),url(canonical),favicon,twitterCard,oEmbedUrl,lang— see the fallbacks above;nullwhen the page does not have them.error—nullon success, otherwiseinvalid_url,blocked_private_address,timeout,dns_failure,connection_error,tls_error,http_404(or anotherhttp_<status>),too_many_redirects,redirect_loop,invalid_redirect_url,not_html(an image, PDF or JSON) orno_preview_found(an HTML page without a title, description or image).- Also:
hasPreview,contentType,charset,bytesRead,truncated(page longer than 1 MB),responseTimeMs,fetchedAt. - SUMMARY record: URLs fetched, previews returned, rows charged, errors by code.
Pricing
Pay per event: $0.0013 per link preview — a URL for which a title, description or image was returned. That is $1.30 per 1,000 previews on the free plan; with Store discounts it is $1.17 (BRONZE), $1.04 (SILVER) and $0.91 per 1,000 (GOLD).
Not charged: rows with an error (fetch errors, 4xx/5xx pages, timeouts, blocked addresses, non-HTML files, pages without any preview data), invalid URLs, and the free demo (empty input).
On the Apify Free plan, the $5 of monthly credit covers about 3,800 link previews a month. Set a maximum charge per run to cap spend: the run stops cleanly when it is reached.
Limits
- Up to 1,000 URLs per run (extra URLs are left out and counted in SUMMARY). Duplicates are removed.
- HTTP only: pages that build their title and Open Graph tags with JavaScript after loading, or that block bots, may return little or nothing — try Website Screenshot for those.
- Only the first 1 MB of HTML is read (Open Graph tags are in the
<head>, so this is plenty). - Polite by default: at most 2 connections per host and at least 0.2 s between requests to the same host. Only fetch URLs you are allowed to access.
FAQ
Is it fast enough for an app? For a handful of URLs, run-sync-get-dataset-items usually answers within a few seconds, most of it the target sites' own response time. For large lists, run it in batches of up to 1,000.
Why is the favicon sometimes /favicon.ico? When the page names no icon, browsers fall back to /favicon.ico at the site root, and so does this Actor. It is not downloaded or checked.
Does it follow redirects? Yes, up to 10, and every hop is checked against private addresses. finalUrl shows where it ended; relative image URLs are resolved against that final page.
Do I pay for broken links? No. Only rows with a preview are charged.
More tools from the same developer
- Bulk URL Status Checker — bulk HTTP status and redirect-chain checker for URL lists
- Redirect Checker — full redirect chain for each URL, hop by hop, with loop detection
- Favicon Extractor — favicons, Apple touch icons and manifest icons of many websites, with the best icon per site
- Website Screenshot — screenshots and PDFs of web pages: full page, mobile, element, no cookie banners
- SEO Audit Tool — on-page SEO audit for URL lists and sitemaps: title, meta description, h1, canonical, noindex, robots.txt, issues and a 0-100 score