Bria

Access Bria's commercial-ready AI models trained exclusively on licensed data through Runware's unified API. Safe for commercial use with built-in content moderation and IP protection.

Introduction

Bria's AI models are integrated into the Runware platform through our unified API, providing access to commercial-ready generative AI trained exclusively on licensed data. Bria specializes in enterprise-safe AI solutions with built-in content moderation, IP protection, and transparency controls designed for professional production environments.

Through the providerSettings.bria object, you can access Bria's unique features such as prompt enhancement, content moderation or IP signal detection, while maintaining the consistency of Runware's standard API structure. This page documents the technical specifications, parameter requirements, and provider-specific settings for all Bria models available through our platform.

Image models

providerSettings » bria
bria
object

Configuration object for Bria-specific features and controls. Bria models offer enterprise-safe AI with built-in content moderation, IP protection, and licensed training data for commercial use.

View example
{
  "taskType": "imageInference",
  "taskUUID": "a770f077-f413-47de-9dac-be0b26a35da6",
  "positivePrompt": "Professional product photography",
  "model": "bria:10@1",
  "width": 1024,
  "height": 1024,
  "providerSettings": {
    "bria": {
      "promptEnhancement": true,
      "enhanceImage": true,
      "contentModeration": true,
      "ipSignal": true
    }
  }
}
Properties ⁨6⁩ properties
providerSettings » bria » promptEnhancement
promptEnhancement
boolean Default: false

Enhances the input prompt with descriptive variations and additional details for more creative and varied outputs. When enabled, the system expands the prompt while maintaining the original intent.

providerSettings » bria » medium
medium
"photography" | "art"

Specifies the artistic medium or style category for generation, influencing the overall aesthetic and rendering approach.

Available values:

  • photography: Optimizes for photorealistic imagery with natural lighting and textures.
  • art: Optimizes for artistic interpretations with stylized rendering and creative expression.
providerSettings » bria » enhanceImage
enhanceImage
boolean Default: false

Generates images with richer details and sharper textures by applying additional enhancement processing to the output. This improves overall visual quality and clarity.

providerSettings » bria » promptContentModeration
promptContentModeration
boolean Default: true

Scans the input prompt for NSFW or restricted terms before generation begins. When enabled, requests with flagged content are rejected before processing to ensure safe commercial use.

providerSettings » bria » contentModeration
contentModeration
boolean Default: true

Applies content moderation to both input visuals and generated outputs, ensuring all content meets safety standards for commercial use. Flagged content results in request rejection.

providerSettings » bria » ipSignal
ipSignal
boolean Default: false

Flags potential intellectual property-related content in the prompt or generated output. When enabled, helps identify potential IP conflicts before commercial use.

Bria 3.2

Commercial-ready text-to-image model with improved aesthetics, strong prompt alignment, and superior short-text rendering, trained exclusively on licensed data for safe commercial use.

Model AIR ID: bria:10@1.

Supported workflows: Text-to-image.

Technical specifications:

  • Positive prompt: 2-3000 characters.
  • Negative prompt: 2-3000 characters (optional).
  • Supported dimensions: 1024×1024 (1:1), 832×1216 (2:3), 1216×832 (3:2), 896×1152 (3:4), 1152×896 (4:3), 896×1088 (4:5), 1088×896 (5:4), 768×1344 (9:16), 1344×768 (16:9).
  • Steps: 4-10 (default: 8).
  • ControlNet models: Supports Bria ControlNet via controlnet parameter. Available models: bria:10@10 (Canny), bria:10@11 (Depth), bria:10@12 (Recoloring), bria:10@13 (Color Grid).
  • IP Adapter: Supports Bria IP Adapter via ipAdapter parameter. Available models: bria:10@20 (Regular), bria:10@21 (Style).

Provider-specific settings:

Parameters supported: promptEnhancement, enhanceImage, medium, promptContentModeration, contentModeration, ipSignal.

{
  "taskType": "imageInference",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d492",
  "model": "bria:10@1",
  "positivePrompt": "Professional product photography of a luxury watch on marble surface",
  "width": 1024,
  "height": 1024,
  "steps": 8,
  "providerSettings": {
    "bria": {
      "promptEnhancement": true,
      "enhanceImage": true,
      "medium": "photography",
      "contentModeration": true
    }
  }
}
{
  "taskType": "imageInference",
  "taskUUID": "6ba7b831-9dad-11d1-80b4-00c04fd430c8",
  "model": "bria:10@1",
  "positivePrompt": "Modern architectural building with clean lines",
  "width": 1152,
  "height": 896,
  "controlnet": [
    {
      "model": "bria:10@10",
      "weight": 0.8,
      "guideImage": "c64351d5-4c59-42f7-95e1-eace013eddab"
    }
  ],
  "providerSettings": {
    "bria": {
      "contentModeration": true
    }
  }
}
{
  "taskType": "imageInference",
  "taskUUID": "550e8400-e29b-41d4-a716-446655440012",
  "model": "bria:10@1",
  "positivePrompt": "Portrait in the same style and lighting",
  "width": 1024,
  "height": 1024,
  "ipAdapters": [
    {
      "model": "bria:10@21",
      "weight": 0.75,
      "guideImage": "c64351d5-4c59-42f7-95e1-eace013eddab"
    }
  ],
  "providerSettings": {
    "bria": {
      "ipSignal": true,
      "contentModeration": true
    }
  }
}

Background removal models

Bria Video Background Removal

Removes complex or moving video backgrounds in real time, isolating subjects with clean alpha channels for seamless compositing and transparency effects.

Model AIR ID: bria:51@1.

Supported workflows: Background removal.

Technical specifications:

  • Input video: Supports inputs.video (required).
  • Input video requirements: Maximum duration 30 seconds, maximum resolution 16000×16000 pixels.
  • Background color: Configurable via settings.rgba parameter. Accepts RGBA color arrays (default: [0, 0, 0, 0] for transparent). Common presets: Black [0, 0, 0, 1], White [255, 255, 255, 1], Gray [128, 128, 128, 1], Red [255, 0, 0, 1], Green [0, 255, 0, 1], Blue [0, 0, 255, 1], Yellow [255, 255, 0, 1], Cyan [0, 255, 255, 1], Magenta [255, 0, 255, 1], Orange [255, 165, 0, 1].
{
  "taskType": "removeBackground",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d494",
  "model": "bria:51@1",
  "inputMedia": "c64351d5-4c59-42f7-95e1-eace013eddab"
}
{
  "taskType": "removeBackground",
  "taskUUID": "6ba7b833-9dad-11d1-80b4-00c04fd430c8",
  "model": "bria:51@1",
  "inputMedia": "c64351d5-4c59-42f7-95e1-eace013eddab",
  "settings": {
    "rgba": [0, 255, 0, 1],
  }
}

Upscaling models

providerSettings » bria
bria
object

Configuration object for Bria-specific upscaling settings. These parameters control how Bria upscaling models handle transparency and image processing.

View example
{
  "taskType": "upscale",
  "taskUUID": "a770f077-f413-47de-9dac-be0b26a35da6",
  "inputs": {
    "image": "c64351d5-4c59-42f7-95e1-eace013eddab",
  },
  "model": "bria:52@1",
  "upscaleFactor": 4,
  "providerSettings": {
    "bria": {
      "preserveAlpha": true
    }
  }
}
Properties ⁨1⁩ property
providerSettings » bria » preserveAlpha
preserveAlpha
boolean Default: true

Controls whether to preserve the original transparency (alpha channel) of the input image during upscaling. When disabled, the output becomes fully opaque.

This parameter only takes effect when the input image contains an alpha channel. Images without transparency are unaffected by this setting.

Bria Image Increase Resolution

Specialized upscaling model that enhances image resolution 2× or 4× while preserving original detail and integrity with support for alpha channel preservation.

Model AIR ID: bria:52@1.

Supported workflows: Image upscaling.

Technical specifications:

  • Input image: Supports inputs.image (required).
  • Upscale factor: 2 or 4 (default: 2).
  • Maximum output resolution: 8192×8192 pixels.
  • Input image requirements: Width and height between 300-2048 pixels, 20MB file size limit.

Provider-specific settings:

Parameters supported: preserveAlpha.

{
  "taskType": "upscale",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d493",
  "model": "bria:52@1",
  "inputImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
  "upscaleFactor": 2
}
{
  "taskType": "upscale",
  "taskUUID": "6ba7b832-9dad-11d1-80b4-00c04fd430c8",
  "model": "bria:52@1",
  "inputImage": "c64351d5-4c59-42f7-95e1-eace013eddab",
  "upscaleFactor": 4,
  "providerSettings": {
    "bria": {
      "preserveAlpha": true
    }
  }
}

Bria Video Increase Resolution

Upscales video resolution by 2× or 4× while preserving detail and quality, supporting multiple output formats and codecs for professional video workflows.

Model AIR ID: bria:50@1.

Supported workflows: Video upscaling.

Technical specifications:

  • Input video: Supports inputs.image (required).
  • Upscale factor: 2 or 4 (default: 2).
  • Maximum output resolution: 7680×4320 (8K).
  • Input video requirements: Maximum 30 seconds duration.
{
  "taskType": "upscale",
  "taskUUID": "f47ac10b-58cc-4372-a567-0e02b2c3d495",
  "model": "bria:50@1",
  "inputMedia": "https://example.com/video.mp4",
  "upscaleFactor": 2
}
{
  "taskType": "upscale",
  "taskUUID": "6ba7b834-9dad-11d1-80b4-00c04fd430c8",
  "model": "bria:50@1",
  "inputMedia": "https://example.com/video.mp4",
  "upscaleFactor": 4
}