ByteDance Seedream 5 Pro Layer Decomposition: AI Image Editor

Decompose complex images into clean, editable multi-layer assets with ByteDance Seedream 5 Pro Layer Decomposition. Try free — pay per generation.

📝

Overview

About this model

ByteDance Seedream 5 Pro Layer Decomposition is an advanced AI model designed to break down raster images into distinct, high-quality component layers. By separating foreground objects, text elements, subject motifs, and background textures into individual PNGs, graphic designers, digital artists, and automated visual workflows can independently edit, reposition, or restyle image elements without manual masking.

1Graphic Design: Instantly extract text and logo overlays from composite artwork.
2Video Production: Isolate foreground characters and background environments for parallax effects.
3UI/UX Prototyping: Separate UI components and illustration layers for interactive web assets.
💰

Pricing & Value

Cost analysis

muapiapp$0.045 per generation (auto/1K/1.5K), $0.09 (2K)

Fast async processing with instant S3 layer hosting.

Fal.aiNot available

Fal.ai currently does not offer the Seedream 5 Pro layer decomposition endpoint.

ReplicateNot available

Replicate currently does not offer the Seedream 5 Pro layer decomposition endpoint.

* Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Promptstring

Optional text description of layer decomposition rules or elements to extract.

Default ValueSplit the content in the image into 4 layers by combining text and subject elements.
Image URLstring

Upload or provide source image URL to decompose into separate layer components.

Default Valuehttps://static.aiquickdraw.com/tools/example/1786019968051_cKRYLHHu.png
ResolutionEnum (4 options)

Target image resolution level.

Default Valueauto
Output FormatEnum (2 options)

File format of generated image layers.

Default Valuepng
📖

Implementation Guide

Developer documentation

How to Use ByteDance Seedream 5 Pro Layer Decomposition

  1. Upload Source Image: Provide the input image URL containing the visual elements or composite scene you wish to decompose.

  2. Configure Decomposition Rules (Optional): Provide a prompt specifying layer isolation criteria, text bounding boxes, or target element breakdown.

  3. Select Size and Format: Choose the output resolution level (auto, 1K, 1.5K, 2K) and output image format (png or jpeg).

API Request Example

curl -X POST "https://api.muapi.ai/api/v1/bytedance-seedream-5.0-pro-layer" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_url": "https://static.aiquickdraw.com/tools/example/1786019968051_cKRYLHHu.png",
    "prompt": "Split the content in the image into separate background and text layers",
    "resolution": "auto",
    "output_format": "png"
  }'

Common Questions

Frequently asked

What does ByteDance Seedream 5 Pro Layer Decomposition do?

ByteDance Seedream 5 Pro Layer Decomposition analyzes a single composite image and separates its visual components (such as text, foreground subjects, and background elements) into clean, standalone image layers.

What output formats are supported?

You can request output layers in transparent PNG or JPEG formats, specified using the output_format parameter.

What resolution levels are available?

The model supports auto, 1K, 1.5K, and 2K resolution levels for high-definition output layers.

Is a text prompt required?

No, a prompt is optional. When omitted, Seedream 5 Pro automatically decomposes the image based on salient visual hierarchy and object boundaries.

How are the generated layers returned?

The API returns an array of image URLs under the output payload, where each item represents an isolated layer.