AI Image Background Remover

Remove the background from images in bulk and get clean cutouts you can use right away.

• Transparent PNG / WebP cutouts of people, products, animals and objects
• Product photos on a solid color: set backgroundColor to #FFFFFF for marketplace-ready white backgrounds, or export JPG
• Crop to subject to trim empty space, and optionally get the black-and-white mask
• Two models: quality (IS-Net, cleaner edges) and fast (U²-Net, about 3x faster)
• Runs open-source models inside the Actor: no third-party API, no subscription, no API key, and your images never leave the run

Typical uses: e-commerce product listings, catalog cleanup, profile photos, design assets, preparing training data, and image pipelines run by AI agents.

Input

{
  "imageUrls": ["https://example.com/product-1.jpg", "https://example.com/product-2.png"],
  "model": "quality",
  "outputFormat": "png",
  "backgroundColor": "",
  "cropToSubject": false,
  "returnMask": false
}

Images must be direct links to JPG, PNG or WebP files, up to 25 MB and 50 megapixels each.

Output (one row per image)

field	example	meaning
outputUrl	https://api.apify.com/v2/key-value-stores/…/records/00000-….png	the cutout, ready to download
imageUrl	https://example.com/product-1.jpg	the original link
width / height	1200 / 900	size of the result
maskUrl	null	the mask, when returnMask is on
error	null	why an image could not be processed

Results are stored in the run's key-value store, so they follow your Apify data retention settings.

Pricing

You pay per successfully processed image, about $5 per 1,000 images. Images that fail to download or open are free, and the mask costs nothing extra. The run stops automatically at your spending limit.

FAQ

Are my images sent anywhere? No. The AI models are bundled inside the Actor, so images are only downloaded from the links you give and processed in the run.

Which model should I use? Start with quality. Use fast for large batches where speed matters more than fine edges such as hair.

Can I process any file? Only process images and documents you own or have permission to use. The Actor does not check rights for you.

How do I call it from code or an AI agent? Use the Apify API or the Apify MCP server with the input shown above; each dataset row links to the finished image.

Credits

Built with rembg (MIT), IS-Net / DIS (Apache-2.0) and U²-Net (Apache-2.0). Not affiliated with their authors.

The sample image in the input form is "Tabby cat with blue eyes" by AdinaVoicu (CC0, via Wikimedia Commons).