Release v18.1.0:

Exposes GPT Lazy Load functionality.

gpt.enableLazyLoad (optional) <Object> - GPT Lazy Load, where configuration parameters allow customization of lazy loading behaviour.

Default settings are

• fetchMarginPercent: 500, // minimum distance from the current viewport a slot must be before we fetch the ad as a percentage of viewport size. 0 means "when the slot enters the viewport", 100 means "when the ad is 1 viewport away"
• renderMarginPercent: 200, // minimum distance from the current viewport a slot must be before we render an ad. This allows for prefetching the ad, but waiting to render and download other subresources. The value works just like fetchMarginPercent as a percentage of viewport.
• mobileScaling: 2.0 // a multiplier applied to margins on mobile devices. This allows varying margins on mobile vs. desktop. For example, a mobileScaling of 2.0 will multiply all margins by 2 on mobile devices, increasing the minimum distance a slot can be before fetching and rendering.