Website Screenshot and PDF Generator

Turn a list of URLs into screenshots or PDFs: the full page or just the visible screen, as a desktop, laptop, tablet or phone would show it. Cookie banners are closed and ads are blocked before the capture, and every file comes back as a download link you can put in a report, an archive or a visual check, or hand to an AI agent that needs to see a page.

Getting a clean capture is the hard part. A screenshot with a cookie banner over half the page, or gray boxes where images should be, isn't much use for before and after shots for a client or proof of how a competitor's page looked on a given day. So the Actor clicks accept on the common consent tools (OneTrust, Cookiebot, Didomi, Usercentrics and more), hides other fixed privacy notices, and scrolls through the page so lazy images load.

What you get for each page

• A PNG, JPEG or PDF file in the run's storage, with a link you can open or download
• The final URL after redirects, the HTTP status and the page title
• File size, load time, and whether a cookie banner was closed
• An error row with the reason for pages that couldn't be loaded, at no cost

Input

{
    "urls": ["https://apify.com", "https://en.wikipedia.org/wiki/Web_scraping"],
    "format": "png",
    "device": "desktop",
    "fullPage": true
}

• format: png, jpeg (smaller files, with jpegQuality) or pdf (A4 or Letter, backgrounds printed).
• device: desktop 1920 x 1080, laptop 1366 x 768, tablet (iPad size) or mobile (iPhone size). Tablet and mobile get a mobile browser, so sites show their mobile layout. viewportWidth and viewportHeight set any size you want.
• fullPage: the whole page top to bottom, or only the first screen.
• selector: capture one element instead of the page, for example #pricing or .hero.
• darkMode asks sites for their dark theme. delayMs waits longer for animations. proxyConfiguration is there for sites that block data centers or change content by country.

Output

{
    "url": "https://www.theverge.com",
    "status": "ok",
    "httpStatus": 200,
    "title": "The Verge",
    "fileUrl": "https://api.apify.com/v2/key-value-stores/.../records/0002-theverge.com.png",
    "format": "png",
    "device": "desktop",
    "bytes": 4232000,
    "loadMs": 2482,
    "bannerClosed": true,
    "cutAtPixels": null
}

The Images view shows the pictures right in the results table.

Pricing

$0.003 per captured page, image or PDF, so 1,000 pages cost $3. Pages that fail to load are free. Apify adds $0.0001 per run for starting the Actor with its default 2 GB of memory. You can set a maximum charge per run and the Actor stops cleanly when it gets there.

Speed

In test runs on 27 September a Wikipedia article loaded in 1 to 2 seconds and The Verge's home page on mobile in 5 to 6. The Actor captures 3 pages at a time, so 100 ordinary pages take a few minutes.

Limits

• Full-page captures stop at 20,000 pixels tall. Very long pages, like endless news feeds, are cut there, and cutAtPixels tells you when that happened.
• Cookie banner handling covers the common consent tools and most home-made notices, not every site. If one stays, bannerClosed is false and a longer delayMs sometimes helps.
• Sites with strong bot protection may show a challenge page instead of the content. Try the residential proxy for those.
• Pages behind a login can't be captured.

Changelog

• 2026-09-27: first version.