Details
The background removal model (RMBG 2.0), released in 2024 and built on the BiRefNet architecture, accepts an image as input and produces a segmented foreground with a pixel-accurate alpha matte as output. The Eraser API accepts an image with a manual or automatic mask and generates contextually coherent fill content to replace removed elements, replicating surrounding lighting and texture. The Image Expansion API extends existing images to new aspect ratios at up to 1-megapixel resolution. RMBG 2.0 is available as open-source for non-commercial use and as a commercially licensed API, SDK, or iFrame. All editing tools are built on Bria's licensed foundation models.
Have evidence about BRIA AI's AI practices? Submit a report.
Report a Sighting →