BRIA RMBG-2.0

Replicate
Image Generation

BRIA's professional background-removal model trained on fully-licensed data. Commercial-safe.

Generate with BRIA RMBG-2.0
Describe what you want and pick a size β€” the image renders inline.
Sign in to try this model with €5 free credits.
Sign in
Result appears here
TL;DRΒ·Last updated May 16, 2026

BRIA RMBG-2.0 is image generation AI model from Replicate, priced at €0.000 per 1M input tokens with a unknown context window.

Try BRIA RMBG-2.0
Sign in to generate β€” 50 free credits on sign-up

Pricing

Price per Generation
Per generation€0.04

API Integration

Use our OpenAI-compatible API to integrate BRIA RMBG-2.0 into your application.

Install
npm install railwail
JavaScript / TypeScript
import railwail from "railwail";

const rw = railwail("YOUR_API_KEY");

const images = await rw.run("bria-rmbg-2-0", "A beautiful sunset over Tokyo");
console.log(images[0].url);

// Or use the image() method for full control
const res = await rw.image("bria-rmbg-2-0", "A cat in space", {
  size: "1024x1024",
  n: 1,
});
console.log(res.data[0].url);
Specifications
Price
€0.04
Developer
Replicate
Category
Image Generation
Supported Formats
image
Tags
bria
image-edit
background-removal
matting
commercial-safe

Deep dive β€” BRIA AI's BRIA RMBG-2.0

About BRIA AI
Founded 2020 Β· Tel Aviv, Israel

BRIA AI was founded in 2020 in Tel Aviv by Yair Adato (CEO) and Vered Horesh, both former senior product and research leaders. The company positions itself as a 'safe-for-commerce' generative-AI platform: every model BRIA ships is trained exclusively on fully licensed and rights-cleared data and includes contractual IP indemnification for customers. BRIA's product line includes proprietary text-to-image models (BRIA 2.x, 3.x), the popular open-research background-removal models RMBG-1.4 and RMBG-2.0, and a suite of image-editing endpoints. BRIA raised a $40M Series B led by GFT Ventures and Intel Capital in 2024 and counts Getty Images, Vidio and many e-commerce brands among its customers.

Visit BRIA AI β†’
Architecture
Encoder-decoder segmentation model (BiRefNet-derived) for binary background segmentation

BRIA RMBG-2.0 (Remove Background 2.0) is the second-generation background-removal model from BRIA AI, released in late 2024 as a successor to RMBG-1.4 (which became one of the most-downloaded models on Hugging Face). RMBG-2.0 is built on the BiRefNet architecture (Zheng et al. 2024, 'Bilateral Reference for High-Resolution Dichotomous Image Segmentation'), an encoder-decoder segmentation network with explicit bilateral reference branches that maintain both low-resolution semantic context and high-resolution boundary detail. The model produces a precise binary alpha matte that segments the foreground subject from the background, supporting hair, fur, glass, transparent objects and complex edges. BRIA trained RMBG-2.0 on a fully licensed dataset of millions of foreground-mask pairs across product photography, portraits, animals and complex industrial scenes. The weights are released for non-commercial research on Hugging Face; commercial use requires a BRIA license including IP indemnification.

Parameters
~220M parameters
Context
unknown
What it can do
  • State-of-the-art binary background removal
  • Excellent handling of hair, fur, glass and transparent objects
  • BiRefNet architecture with bilateral reference branches
  • Fast inference (~50-200ms on GPU)
  • Trained on fully licensed data β€” safe-for-commerce
  • BRIA offers IP indemnification under commercial license
  • Used by e-commerce, advertising and media companies at scale
  • Best for: e-commerce product cutouts, portrait isolation, pre-processing for compositing pipelines.
Training & License

Trained on millions of fully licensed foreground-mask pairs curated by BRIA AI across product photography, portraits, animals, vehicles and industrial scenes.

License: Open weights on Hugging Face under BRIA RMBG-2.0 license β€” non-commercial use permitted; commercial use requires a BRIA enterprise license that includes IP indemnification.

Known limitations
  • Binary alpha only β€” no trimap or detailed matting nuance
  • Struggles on very fine hair against busy backgrounds
  • Open weights restricted to non-commercial use
  • Single subject focus β€” multi-subject scenes need additional logic

Frequently asked questions

Start using BRIA RMBG-2.0 today

Get started with free credits. No credit card required. Access BRIA RMBG-2.0 and 100+ other models through a single API.