Long webpage screenshots and where the limits actually are

Short answer

A long webpage screenshot works the same way as any full page capture, but past roughly 16,384 pixels of height it hits a hard browser limit on how large a single image can be. Beyond that the capture has to be split into parts that tile the page exactly, with no gap and no overlap between them.

Updated 13 August 2026Written by the Full Page Screenshot team

Most pages are a few thousand pixels tall and nothing interesting happens. Documentation sites, long-form articles, chat exports and analytics dashboards are a different category: 20,000, 40,000, occasionally 100,000 pixels. At that scale you stop fighting the page and start fighting the browser.

This page is about that second category — what breaks, why, and what a capture can honestly promise.

The limits, precisely

These are properties of the browser's graphics layer, not of any particular extension. No tool can raise them.

Maximum canvas edge (Chrome, desktop): 16,384 px
Maximum canvas area: About 268 million px²
Height at 1,440 px wide before splitting: About 16,384 px
Effective height on a 2× display: Halved — density counts
What happens past the limit: Capture splits into tiled parts
Overlap between parts: None. Parts join edge to edge.

What goes wrong on very tall pages

Silent truncation: The worst failure mode: a tool hits the canvas limit, the draw operation fails quietly, and you get an image that looks complete but stops partway down. You only notice when you go looking for something that is missing. Splitting into parts is the honest alternative.
Memory exhaustion: Holding a 40,000 pixel bitmap in memory while also holding each incoming capture is how a tab crashes. Encoding and releasing bands as they complete keeps peak memory roughly constant regardless of page height.
Rate limiting: Chrome rate-limits how often a tab can be photographed. A 40,000 pixel page needs around fifty captures, so the capture has to detect the limit and back off rather than dropping frames.
The page changing underneath you: A long capture takes several seconds. Lazy content, sticky elements re-attaching, and layout shifts all happen during that window, which is why the page state has to be pinned at the start rather than assumed.

When a long screenshot is the wrong answer

A 40,000 pixel PNG is awkward. Image viewers scale it to unreadable, email clients reject the attachment, and printing it is hopeless.

For anything longer than about 15,000 pixels, exporting to PDF is usually better: the content is paginated at a readable size, page breaks land in the gaps between lines rather than through them, and the file opens in anything. The image is right when you need one continuous canvas; the PDF is right when you need a document someone will read.

Frequently asked questions

How long a page can be captured?

There is no page-length limit on the capture itself. Past about 16,384 pixels the output is delivered as multiple parts rather than one image, because that is the largest single image a browser can hold.

Do the parts overlap or lose content between them?

No. Parts tile the image exactly — the last row of one part is immediately followed by the first row of the next.

Why is my long screenshot cut off at the bottom?

Usually the canvas limit being hit silently by whatever tool produced it. It can also be an infinite-scroll page that had not finished loading when the capture started.

Is there a faster way to capture a very long page?

Reducing the browser zoom before capturing fits more content into each screenful, which means fewer captures and a smaller final image, at the cost of rendered detail.

Capture your first full page screenshot

Free, no account, and nothing leaves your computer.

Add to Chrome — Free Read the guide

Keep reading