Bria Background API

Remove
Replace

Endpoint: POST https://fal.run/fal-ai/bria/background/remove Endpoint ID: fal-ai/bria/background/remove

Try it in the Playground

Run this model interactively with your own prompts.

Quick Start

import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "fal-ai/bria/background/remove",
    arguments={
        "image_url": "https://fal.media/files/panda/K5Rndvzmn1j-OI1VZXDVd.jpeg"
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)

import { fal } from "@fal-ai/client";

const result = await fal.subscribe("fal-ai/bria/background/remove", {
  input: {
      image_url: "https://fal.media/files/panda/K5Rndvzmn1j-OI1VZXDVd.jpeg"
    },
  logs: true,
  onQueueUpdate: (update) => {
    if (update.status === "IN_PROGRESS") {
      update.logs.map((log) => log.message).forEach(console.log);
    }
  },
});
console.log(result.data);
console.log(result.requestId);

curl --request POST \
  --url https://fal.run/fal-ai/bria/background/remove \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "image_url": "https://fal.media/files/panda/K5Rndvzmn1j-OI1VZXDVd.jpeg"
}'

Examples

Input Schema

string

required

Input Image to erase from

boolean

default:"false"

If True, the media will be returned as a data URI and the output data won’t be available in the request history.

Output Schema

Image

required

The generated image

Input Example

{
  "image_url": "https://fal.media/files/panda/K5Rndvzmn1j-OI1VZXDVd.jpeg",
  "sync_mode": false
}

Output Example

{
  "image": {
    "content_type": "image/png",
    "file_name": "070c731993e949d993c10ef6283d335d.png",
    "file_size": 1076276,
    "height": 1024,
    "url": "https://v3.fal.media/files/tiger/GQEMNjRyxSoza7N8LPPqb_070c731993e949d993c10ef6283d335d.png",
    "width": 1024
  }
}

Endpoint: POST https://fal.run/fal-ai/bria/background/replace Endpoint ID: fal-ai/bria/background/replace

Try it in the Playground

Run this model interactively with your own prompts.

Quick Start

import fal_client

def on_queue_update(update):
    if isinstance(update, fal_client.InProgress):
        for log in update.logs:
           print(log["message"])

result = fal_client.subscribe(
    "fal-ai/bria/background/replace",
    arguments={
        "image_url": "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_fg.jpg"
    },
    with_logs=True,
    on_queue_update=on_queue_update,
)
print(result)

import { fal } from "@fal-ai/client";

const result = await fal.subscribe("fal-ai/bria/background/replace", {
  input: {
      image_url: "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_fg.jpg"
    },
  logs: true,
  onQueueUpdate: (update) => {
    if (update.status === "IN_PROGRESS") {
      update.logs.map((log) => log.message).forEach(console.log);
    }
  },
});
console.log(result.data);
console.log(result.requestId);

curl --request POST \
  --url https://fal.run/fal-ai/bria/background/replace \
  --header "Authorization: Key $FAL_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "image_url": "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_fg.jpg"
}'

Input Schema

string

required

Input Image to erase from

string

default:""

The URL of the reference image to be used for generating the new background. Use "" to leave empty. Either ref_image_url or bg_prompt has to be provided but not both. If both ref_image_url and ref_image_file are provided, ref_image_url will be used. Accepted formats are jpeg, jpg, png, webp. Default value: ""

string

The prompt you would like to use to generate images.

string

default:""

The negative prompt you would like to use to generate images. Default value: ""

boolean

default:"true"

Whether to refine prompt Default value: true

integer

The same seed and the same prompt given to the same version of the model will output the same image every time.Range: 0 to 2147483647

boolean

default:"true"

Whether to use the fast model Default value: true

integer

default:"1"

Number of Images to generate. Default value: 1Range: 1 to 4

boolean

default:"false"

If True, the media will be returned as a data URI and the output data won’t be available in the request history.

Output Schema

list<Image>

required

The generated images

integer

required

Seed value used for generation.

Input Example

{
  "image_url": "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_fg.jpg",
  "ref_image_url": "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_bg.jpg",
  "prompt": "Man leaning against a wall",
  "negative_prompt": "",
  "refine_prompt": true,
  "fast": true,
  "num_images": 1,
  "sync_mode": false
}

Output Example

{
  "images": [
    {
      "content_type": "image/png",
      "url": "https://storage.googleapis.com/falserverless/bria/bria_bg_replace_res.jpg"
    }
  ]
}

Bria’s RMBG 2.0 delivers production-grade background removal at $0.018 per inference, trained exclusively on licensed commercial data. Trading breadth for legal certainty, this model solves the compliance headache that blocks most background removal tools from enterprise deployment. Built for teams that need defensible image segmentation without licensing risk. Use Cases: E-commerce Product Photography | Marketing Asset Production | Professional Photo Editing

Performance

At $0.018 per generation, RMBG 2.0 is a specialized utility model, comparable to standard background removal pricing but with the critical differentiator of commercial licensing clarity.

Metric	Result	Context
Training Data	100% licensed commercial imagery	Legal defensibility for enterprise deployment
Cost per Image	$0.018	55 generations per $1.00 on fal
Input Format	Single image URL	Standard image-to-image pipeline
Output Format	PNG with transparency	1024x1024 preserved dimensions
Related Endpoints	Background Replace, Product Shot, Expand Image	Same-family variants for product photography workflows

Built for Commercial Deployment

RMBG 2.0 uses a segmentation architecture trained exclusively on licensed data, contrasting sharply with open models trained on scraped web imagery. What this means for you:

Enterprise-safe training data: Every training image carries commercial licensing, eliminating copyright exposure in regulated industries
Transparent background removal: Clean PNG output with alpha channel preservation for immediate compositing workflows
Single-purpose optimization: Focused solely on background segmentation rather than multi-task compromise
Straightforward integration: Standard image URL input via fal’s API with synchronous or asynchronous processing modes

Technical Specifications

Spec	Details
Architecture	Bria RMBG 2.0
Input Formats	JPEG, PNG, WebP, GIF, AVIF via URL
Output Formats	PNG with alpha channel transparency
Resolution	Up to 1024x1024 (preserves input dimensions)
License	Commercial use with licensed training data

API Documentation | Quickstart Guide | Enterprise Pricing

How It Stacks Up

**Bria Background Replace (

0.023)** – RMBG 2.0 (

0.018) strips backgrounds cleanly, 22% more cost-effective. Background Replace adds generative fill capabilities for product staging workflows where you need both removal and context reconstruction. **Bria Product Shot (

0.023)** – RMBG 2.0 (

0.018) handles pure segmentation, while Product Shot combines removal with studio-quality scene generation for e-commerce listings. Product Shot trades simplicity for complete product photography automation at 28% higher cost. **Bria Expand Image (

0.023)** – RMBG 2.0 focuses on isolation; [Expand Image](https://fal.ai/models/fal-ai/bria/expand) handles outpainting for aspect ratio adjustments post-removal. Teams typically chain RMBG 2.0 into Expand for complete product asset workflows, with Expand adding generative borders at

0.023 per operation.

Bria Expand Image — Image Generation
Bria Eraser — Image Generation
Bria Product Shot — Image Generation
Bria Background Replace — Image Generation
Video — Video Generation
Bria GenFill — Image Generation
Replace Background — Image Generation
Bria Text-to-Image HD — Image Generation
Bria Text-to-Image Base — Image Generation
Embed Product — Image Generation
Bria Text-to-Image Fast — Image Generation
Bria RMBG 2.0 — Image Generation

Limitations

seed range: 0 to 2147483647
num_images range: 1 to 4

Video Generation

Image Generation

Audio

Vision

3D

Bria Background API

Try it in the Playground

Quick Start

Examples

Input Schema

Output Schema

Input Example

Output Example

Try it in the Playground

Quick Start

Input Schema

Output Schema

Input Example

Output Example

Performance

Built for Commercial Deployment

Technical Specifications

How It Stacks Up

Limitations

Try it in the Playground

​Quick Start

​Examples

​Input Schema

​Output Schema

​Input Example

​Output Example

Try it in the Playground

​Quick Start

​Input Schema

​Output Schema

​Input Example

​Output Example

​Performance

​Built for Commercial Deployment

​Technical Specifications

​How It Stacks Up

​Related

​Limitations

Quick Start

Examples

Input Schema

Output Schema

Input Example

Output Example

Quick Start

Input Schema

Output Schema

Input Example

Output Example

Performance

Built for Commercial Deployment

Technical Specifications

How It Stacks Up

Related

Limitations