Available now

BiRefNet Background Removal: background removal

BiRefNet separates the subject of an image from its background and returns the cut-out, with the mask too if you want it.

Model ID

Input

Fields marked * are required. Review the price before generating.

The image to remove background from

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

Additional settings

The resolution to operate on. The higher the resolution, the more accurate the output will be for high res input images.

Whether to output the mask used to remove the background

Whether to refine the foreground using the estimated mask

Price$0.28 / run

You sign up first. Your settings are kept.

About this model

Overview

Upload an image and BiRefNet removes the background. Choose between a light general model, a heavier general model that is slower and more accurate, and a model trained for portraits.

It works at 1024x1024 or 2048x2048; the larger size helps with high resolution photos. It can also return the mask it used, and saves the result as PNG, WebP or GIF.

What you can build

  • Product cut-outs for listings and catalogs
  • Portrait cut-outs for profile photos and team pages
  • Black and white masks for compositing in another editor
  • Transparent layers for banners and collages

Prompting tips

  • Start with the light general model; switch to the heavy one when edges such as fur or thin parts come out rough.
  • Pick the portrait model for photos of people.
  • Turn on the mask output when you want to finish the cut-out yourself in an image editor.

What to check

Check hair, transparent objects and low-contrast edges over different backgrounds. Foreground refinement can change edge detail; compare it with the original before compositing.

Inputs

  • Model (optional): Model to use for background removal. The 'General Use (Light)' model is the original model used in the BiRefNet repository. The 'General Use (Heavy)' model is a slower but more accurate model. The 'Portrait' model is a model trained specifically for portrait images. The 'General Use (Light)' model is recommended for most use cases. Options: General Use (Light), General Use (Heavy), Portrait.
  • Operating Resolution (optional): The resolution to operate on. The higher the resolution, the more accurate the output will be for high res input images. Options: 1024x1024, 2048x2048.
  • Output Mask (optional): Whether to output the mask used to remove the background.
  • Refine Foreground (optional): Whether to refine the foreground using the estimated mask.
  • Image (required): URL of the image to remove background from.
  • Output Format (optional): The format of the output image. Options: webp, png, gif.

Output and limits

  • File formats: WEBP, PNG, GIF.
  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • up to: $0.28 / run