Website Screenshot API: Full-Page & Bulk

Get the full scrolled page, not just the top of it (up to 8,000 px tall by default; raise maxHeight to 30,000 for longer pages). This actor scrolls each page to load lazy images, keeps sticky headers from repeating down the shot, hides many common cookie banners (some sites' own consent walls may still show), and renders emoji and Asian-language text properly. Send a list of URLs and it captures them in parallel, as PNG, JPEG or PDF, with options for width, height, device presets, a single element or a fixed crop. You pay $5 per 1,000 successful screenshots ($0.005 each), plus $0.01 for each very large image (taller than 8,000 px, or a PNG taller than 4,000 px). There's no start fee, and failed and fully blank pages are free. Only urls is required; every other setting has a sensible default.

What you get

• Devices: desktop (1920×1080), laptop (1366×768), tablet or mobile, or your own width and height.
• Full page or visible area. You can also capture just the top N pixels, or just one element by CSS selector (e.g. #pricing).
• Formats: JPEG (default) or PNG, plus an optional PDF of the page.
• Clean shots: many common cookie banners hidden, fixed headers and chat widgets fixed, or hide any elements you list. Dark mode is available.
• Emoji and CJK fonts render correctly (no empty boxes).
• Speed: one browser with parallel tabs. Short pages take a few seconds; long, image-heavy pages (10,000+ px tall) can take 1–2 minutes each at 2 GB memory.
• Sizes: mobile and tablet render at 2× pixel density; maxHeight is in output pixels and the width never exceeds the device width.
• Hard time limit per URL (2 × timeoutSecs + 30 s), so one stuck page can't hang your run.

Input example

{ "urls": ["https://apify.com", "stripe.com"], "device": "desktop", "fullPage": true, "format": "jpeg" }

Output

One dataset row per screenshot, with:

• screenshotUrl (a direct image link);
• title, status, pageHeight, truncated, bytes, ms;
• pdfUrl if you asked for a PDF.

The images are stored in the run's key-value store.

The SUMMARY record lists any URLs that failed and why. They aren't charged.

Pricing

$0.005 per successful screenshot ($5 per 1,000). There's no start fee, and failed and fully blank pages are free.

Large captures: +$0.01 each, so $0.015 in total for that screenshot. Charged when the saved image is taller than 8,000 px, or is a PNG taller than 4,000 px. This can happen when you raise maxHeight, pick PNG for a long page, set a tall clipHeight, or capture a tall element. The default (JPEG, full page cut at 8,000 px) never adds it. The row shows largeCapture: true.

Set "Maximum cost per run" to cap spend.

Tips and limits

• Very long pages are cut at maxHeight (8,000 px by default; set it up to 30,000 for the true full page; images over 8,000 px add the $0.01 large-capture charge). The row shows truncated: true.
• Pages behind logins, or that block automated browsers, may fail. Those are free and listed in SUMMARY.
• Not right? Try waitUntil: networkidle or a longer delayMs for slow pages.
• Questions or feature requests: open an Issue on this page, or email mystiquelabinfo@gmail.com.