bytedance/seedream-v5.0-flash/layer-decomposition

image-to-image

Seedream v5.0 Flash Layer Decomposition API by ByteDance

bytedance/seedream-v5.0-flash/layer-decomposition

Layer-decomposition

Dola-Seedream-5.0-flash layer decomposition from ByteDance's high-speed, cost-optimized model built for instant creation. Splits one image into a base image plus up to 16 transparent PNG layers with stacking order (z_index), bounding boxes, names, and descriptions - fast, bulk-scale design-asset extraction.

Seedream v5.0 Flash Layer Decomposition is developed by ByteDance. Atlas Cloud (operated by Atlas Cloud AI LLC) provides access to it and does not own it. All trademarks belong to their respective owners.

INPUT

Prompt

Image *

You can drag and drop a file or click to upload

MAX:1

Size

Output format

OUTPUT

Idle

Your generated images will appear here

Configure your settings and click Run to get started

Your request will cost $0.018 per run. For $10 you can run this model approximately 555 times.Billed by the actual number of images returned (base + layers). The estimate shows the maximum.

Here's what you can do next:

Image-to-Video Image-to-Image

Parameters

Code Example
import requests
import time

# Step 1: Start image generation
generate_url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
    "model": "bytedance/seedream-v5.0-flash/layer-decomposition",  # Required. Model identifier
    "prompt": "A beautiful landscape with mountains and lake",  # Optional instruction describing which elements to split into layers
    "image": "example_value",  # Required. The image to decompose into layers
    "size": "auto",  # Output resolution tier. options: auto | 1K | 1.5K | 2K
    "output_format": "jpeg",  # File format of the base image only. options: jpeg | png
    "enable_sync_mode": False,  # If set to true, the function will wait for the result to be generated and uploaded before returning the response
    "enable_base64_output": False,  # If enabled, the output will be encoded into a BASE64 string instead of a URL
}

generate_response = requests.post(generate_url, headers=headers, json=data)
generate_result = generate_response.json()
prediction_id = generate_result["data"]["id"]

# Step 2: Poll for result
poll_url = f"https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}"

def check_status():
    while True:
        response = requests.get(poll_url, headers={"Authorization": "Bearer $ATLASCLOUD_API_KEY"})
        result = response.json()

        if result["data"]["status"] == "completed":
            print("Generated image:", result["data"]["outputs"][0])
            return result["data"]["outputs"][0]
        elif result["data"]["status"] == "failed":
            raise Exception(result["data"]["error"] or "Generation failed")
        else:
            # Still processing, wait 2 seconds
            time.sleep(2)

image_url = check_status()

Install

Install the required package for your language.

pip install requests

Authentication

All API requests require authentication via an API key. You can get your API key from the Atlas Cloud dashboard.

export ATLASCLOUD_API_KEY="your-api-key-here"

HTTP Headers

import os

API_KEY = os.environ.get("ATLASCLOUD_API_KEY")
headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {API_KEY}"
}

Keep your API key secure

Never expose your API key in client-side code or public repositories. Use environment variables or a backend proxy instead.

Submit a request

import requests

url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
    "model": "your-model",
    "prompt": "A beautiful landscape"
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

Submit a Request

Submit an asynchronous generation request. The API returns a prediction ID that you can use to check the status and retrieve the result.

POST/api/v1/model/generateImage

Request Body

import requests

url = "https://api.atlascloud.ai/api/v1/model/generateImage"
headers = {
    "Content-Type": "application/json",
    "Authorization": "Bearer $ATLASCLOUD_API_KEY"
}

data = {
    "model": "bytedance/seedream-v5.0-flash/layer-decomposition",
    "prompt": "A beautiful landscape with mountains and lake"
}

response = requests.post(url, headers=headers, json=data)
result = response.json()

print(f"Prediction ID: {result['data']['id']}")
print(f"Status: {result['data']['status']}")

Response

{
  "code": 200,
  "data": {
    "id": "pred_abc123",
    "status": "processing",
    "model": "model-name",
    "created_at": "2025-01-01T00:00:00Z"
  }
}

Check Status

Poll the prediction endpoint to check the current status of your request.

GET/api/v1/model/prediction/{prediction_id}

Polling Example

import requests
import time

prediction_id = "pred_abc123"
url = f"https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}"
headers = { "Authorization": "Bearer $ATLASCLOUD_API_KEY" }

while True:
    response = requests.get(url, headers=headers)
    result = response.json()
    status = result["data"]["status"]
    print(f"Status: {status}")

    if status in ["completed", "succeeded"]:
        output_url = result["data"]["outputs"][0]
        print(f"Output URL: {output_url}")
        break
    elif status == "failed":
        print(f"Error: {result['data'].get('error', 'Unknown')}")
        break

    time.sleep(3)

Status Values

processingThe request is still being processed.

completedGeneration is complete. Outputs are available.

succeededGeneration succeeded. Outputs are available.

failedGeneration failed. Check the error field.

Completed Response

{
  "data": {
    "id": "pred_abc123",
    "status": "completed",
    "outputs": [
      "https://storage.atlascloud.ai/outputs/result.png"
    ],
    "metrics": {
      "predict_time": 8.3
    },
    "created_at": "2025-01-01T00:00:00Z",
    "completed_at": "2025-01-01T00:00:10Z"
  }
}

Upload Files

Upload files to Atlas Cloud storage and get a URL you can use in your API requests. Use multipart/form-data to upload.

POST/api/v1/model/uploadMedia

Upload Example

import requests

url = "https://api.atlascloud.ai/api/v1/model/uploadMedia"
headers = { "Authorization": "Bearer $ATLASCLOUD_API_KEY" }

with open("image.png", "rb") as f:
    files = {"file": ("image.png", f, "image/png")}
    response = requests.post(url, headers=headers, files=files)

result = response.json()
download_url = result["data"]["download_url"]
print(f"File URL: {download_url}")

Response

{
  "data": {
    "download_url": "https://storage.atlascloud.ai/uploads/abc123/image.png",
    "file_name": "image.png",
    "content_type": "image/png",
    "size": 1024000
  }
}

Input Schema

The following parameters are accepted in the request body.

Total: 7Required: 2Optional: 5

modelstringrequired

Model identifier.

Default: "bytedance/seedream-v5.0-flash/layer-decomposition"

promptstring

Optional instruction describing which elements to split into layers. Leave empty to let the model automatically detect and split all major elements. You can describe targets in natural language, or pinpoint them with <bbox>x1 y1 x2 y2</bbox> tags using normalized [0, 1000] coordinates (origin at the top-left corner).

Example: "Split the poster into layers: the title text <bbox>179 58 809 197</bbox> and the parrot <bbox>330 274 641 991</bbox>."

imagestringrequired

The image to decompose into layers. Accepts a public URL or Base64-encoded image (png, jpeg, webp, bmp, tiff, gif). Exactly one image; total pixels must be within [512x512, 6000x6000] and aspect ratio within [1/16, 16], max 30MB.

sizestring

Output resolution tier. auto follows the input image's size (capped to the 1K-2K range). The base image keeps the input's aspect ratio; each layer keeps its own aspect ratio from the original.

Default: "auto"

auto1K1.5K2K

output_formatstring

File format of the base image only. Decomposed layers are always transparent png.

Default: "jpeg"

jpegpng

enable_sync_modeboolean

If set to true, the function will wait for the result to be generated and uploaded before returning the response. It allows you to get the result directly in the response. This property is only available through the API.

Default: false

enable_base64_outputboolean

If enabled, the output will be encoded into a BASE64 string instead of a URL. This property is only available through the API.

Default: false

Example Request Body

{
  "model": "bytedance/seedream-v5.0-flash/layer-decomposition",
  "image": "example_image",
  "size": "auto",
  "output_format": "jpeg",
  "enable_sync_mode": false,
  "enable_base64_output": false
}

Output Schema

The API returns a prediction response with the generated output URLs.

codeinteger

Status code of the call; 200 on success.

messagestring

Status message of the call.

dataobject

The prediction: id, model, status, outputs (base image first, then each layer in ascending z-order) and index-aligned layers metadata, plus urls and created_at. Poll GET /api/v1/model/prediction/{request_id} with data.id until status is completed or failed.

Example Response

{
  "id": "pred_abc123",
  "status": "completed",
  "model": "model-name",
  "outputs": [
    "https://storage.atlascloud.ai/outputs/result.png"
  ],
  "metrics": {
    "predict_time": 8.3
  },
  "created_at": "2025-01-01T00:00:00Z",
  "completed_at": "2025-01-01T00:00:10Z"
}

Atlas Cloud Skills

Atlas Cloud Skills integrates 400+ AI models directly into your AI coding assistant. One command to install, then use natural language to generate images, videos, and chat with LLMs.

Supported Clients

Claude Code

OpenAI Codex

Gemini CLI

Cursor

Windsurf

VS Code

Trae

GitHub Copilot

Cline

Roo Code

Amp

Goose

Replit

40+ supported clients

Install

npx skills add AtlasCloudAI/atlas-cloud-skills

Setup API Key

Get your API key from the Atlas Cloud dashboard and set it as an environment variable.

export ATLASCLOUD_API_KEY="your-api-key-here"

Capabilities

Once installed, you can use natural language in your AI assistant to access all Atlas Cloud models.

Image GenerationGenerate images with models like Nano Banana 2, Z-Image, and more.

Video CreationCreate videos from text or images with Kling, Vidu, Veo, etc.

LLM ChatChat with Qwen, DeepSeek, and other large language models.

Media UploadUpload local files for image editing and image-to-video workflows.

Learn more

github.com/AtlasCloudAI/atlas-cloud-skills

MCP Server

Atlas Cloud MCP Server connects your IDE with 400+ AI models via the Model Context Protocol. Works with any MCP-compatible client.

Supported Clients

Cursor

VS Code

Windsurf

Claude Code

OpenAI Codex

Gemini CLI

Cline

Roo Code

100+ supported clients

Install

npx -y atlascloud-mcp

Configuration

Add the following configuration to your IDE's MCP settings file.

{
  "mcpServers": {
    "atlascloud": {
      "command": "npx",
      "args": [
        "-y",
        "atlascloud-mcp"
      ],
      "env": {
        "ATLASCLOUD_API_KEY": "your-api-key-here"
      }
    }
  }
}

Available Tools

atlas_generate_imageGenerate images from text prompts.

atlas_generate_videoCreate videos from text or images.

atlas_chatChat with large language models.

atlas_list_modelsBrowse 400+ available AI models.

atlas_quick_generateOne-step content creation with auto model selection.

atlas_upload_mediaUpload local files for API workflows.

Learn more

github.com/AtlasCloudAI/mcp-server

API Schema

{
  "openapi": "3.0.0",
  "info": {
    "title": "AtlasCloud API - bytedance/seedream-v5.0-flash/layer-decomposition",
    "version": "1.0.0",
    "description": "The AtlasCloud API."
  },
  "servers": [
    {
      "url": "https://api.atlascloud.ai"
    }
  ],
  "paths": {
    "/api/v1/model/generateImage": {
      "post": {
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/Input"
              }
            }
          }
        },
        "responses": {
          "200": {
            "description": "The request status.",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/PredictionResponse"
                }
              }
            }
          }
        }
      },
      "x-api-name": "model_run"
    },
    "/api/v1/model/prediction/{request_id}": {
      "get": {
        "parameters": [
          {
            "in": "path",
            "name": "request_id",
            "required": true,
            "schema": {
              "type": "string",
              "description": "Request ID"
            }
          }
        ],
        "responses": {
          "200": {
            "description": "Result of the request.",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/PredictionResponse"
                }
              }
            }
          }
        }
      },
      "x-api-name": "model_result"
    }
  },
  "components": {
    "schemas": {
      "Input": {
        "type": "object",
        "properties": {
          "model": {
            "type": "string",
            "default": "bytedance/seedream-v5.0-flash/layer-decomposition",
            "description": "Model identifier."
          },
          "prompt": {
            "type": "string",
            "description": "Optional instruction describing which elements to split into layers. Leave empty to let the model automatically detect and split all major elements. You can describe targets in natural language, or pinpoint them with <bbox>x1 y1 x2 y2</bbox> tags using normalized [0, 1000] coordinates (origin at the top-left corner).",
            "example": "Split the poster into layers: the title text <bbox>179 58 809 197</bbox> and the parrot <bbox>330 274 641 991</bbox>."
          },
          "image": {
            "type": "string",
            "description": "The image to decompose into layers. Accepts a public URL or Base64-encoded image (png, jpeg, webp, bmp, tiff, gif). Exactly one image; total pixels must be within [512x512, 6000x6000] and aspect ratio within [1/16, 16], max 30MB."
          },
          "size": {
            "type": "string",
            "enum": [
              "auto",
              "1K",
              "1.5K",
              "2K"
            ],
            "default": "auto",
            "x-ui-component": "select",
            "description": "Output resolution tier. auto follows the input image's size (capped to the 1K-2K range). The base image keeps the input's aspect ratio; each layer keeps its own aspect ratio from the original."
          },
          "output_format": {
            "type": "string",
            "enum": [
              "jpeg",
              "png"
            ],
            "default": "jpeg",
            "description": "File format of the base image only. Decomposed layers are always transparent png."
          },
          "enable_sync_mode": {
            "type": "boolean",
            "default": false,
            "disabled": true,
            "description": "If set to true, the function will wait for the result to be generated and uploaded before returning the response. It allows you to get the result directly in the response. This property is only available through the API."
          },
          "enable_base64_output": {
            "type": "boolean",
            "default": false,
            "disabled": true,
            "description": "If enabled, the output will be encoded into a BASE64 string instead of a URL. This property is only available through the API."
          }
        },
        "required": [
          "model",
          "image"
        ],
        "x-order-properties": [
          "model",
          "prompt",
          "image",
          "size",
          "output_format",
          "enable_sync_mode",
          "enable_base64_output"
        ]
      },
      "PredictionResponse": {
        "type": "object",
        "properties": {
          "code": {
            "type": "integer",
            "description": "Status code of the call; 200 on success."
          },
          "message": {
            "type": "string",
            "description": "Status message of the call."
          },
          "data": {
            "type": "object",
            "description": "The prediction: id, model, status, outputs (base image first, then each layer in ascending z-order) and index-aligned layers metadata, plus urls and created_at. Poll GET /api/v1/model/prediction/{request_id} with data.id until status is completed or failed.",
            "allOf": [
              {
                "$ref": "#/components/schemas/PredictionData"
              }
            ]
          }
        }
      },
      "PredictionData": {
        "type": "object",
        "properties": {
          "id": {
            "type": "string",
            "description": "Unique identifier for the prediction."
          },
          "model": {
            "type": "string",
            "description": "Model ID used for the prediction."
          },
          "status": {
            "type": "string",
            "description": "Status of the task: created, processing, completed, or failed."
          },
          "outputs": {
            "type": "array",
            "items": {
              "type": "string"
            },
            "description": "URLs of the generated images: the base image first, then each decomposed layer in ascending z-order (up to 1 base + 16 layers). Empty when status is not completed."
          },
          "layers": {
            "type": "array",
            "description": "Per-output layer metadata, index-aligned with outputs (layers[i] describes outputs[i]). The base image entry has z_index 0 and no bounding box, name, or description.",
            "items": {
              "type": "object",
              "properties": {
                "z_index": {
                  "type": "integer",
                  "description": "Stacking order from the bottom up: 0 is the base image; larger values sit higher. Compositing layers by ascending z_index inside their bounding boxes reconstructs the full image."
                },
                "bounding_box": {
                  "type": "object",
                  "description": "The region this layer occupies within the base image. Absent on the base image entry.",
                  "properties": {
                    "absolute": {
                      "type": "array",
                      "items": {
                        "type": "integer"
                      },
                      "description": "[x1, y1, x2, y2] pixel coordinates of the top-left and bottom-right corners in the output base image's coordinate system (origin at top-left, x right, y down)."
                    },
                    "normalized": {
                      "type": "array",
                      "items": {
                        "type": "integer"
                      },
                      "description": "[x1, y1, x2, y2] quantized to the [0, 1000] range, proportional to the base image size."
                    }
                  }
                },
                "name": {
                  "type": "string",
                  "description": "Model-generated label of the layer's subject. Absent on the base image entry."
                },
                "description": {
                  "type": "string",
                  "description": "Model-generated description of the layer (color, state, material, ...). Absent on the base image entry."
                }
              }
            }
          },
          "urls": {
            "type": "object",
            "description": "Object containing related API endpoints."
          },
          "created_at": {
            "type": "string",
            "format": "date-time",
            "description": "ISO timestamp of when the request was created."
          }
        }
      }
    },
    "securitySchemes": {
      "apiKeyAuth": {
        "type": "apiKey",
        "in": "header",
        "name": "Authorization"
      }
    }
  }
}

LLM-Friendly Prompt Template

# bytedance/seedream-v5.0-flash/layer-decomposition

> Dola-Seedream-5.0-flash layer decomposition from ByteDance's high-speed, cost-optimized model built for instant creation. Splits one image into a base image plus up to 16 transparent PNG layers with stacking order (z_index), bounding boxes, names, and descriptions - fast, bulk-scale design-asset extraction.


## Overview

- **Submit endpoint (POST)**: `https://api.atlascloud.ai/api/v1/model/generateImage` — start an async generation; returns a `prediction_id`
- **Poll endpoint (GET)**: `https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}` — poll this until the prediction finishes
- **Model ID**: `bytedance/seedream-v5.0-flash/layer-decomposition`


## API Information

This model can be used via our HTTP API or more conveniently via our client libraries.
See the input and output schema below, as well as the usage examples.


### Input Schema

The API accepts the following input parameters:

- **`model`** (`string`, _required_):
  Model identifier.
  - Default: `"bytedance/seedream-v5.0-flash/layer-decomposition"`

- **`prompt`** (`string`, _optional_):
  Optional instruction describing which elements to split into layers. Leave empty to let the model automatically detect and split all major elements. You can describe targets in natural language, or pinpoint them with <bbox>x1 y1 x2 y2</bbox> tags using normalized [0, 1000] coordinates (origin at the top-left corner).

- **`image`** (`string`, _required_):
  The image to decompose into layers. Accepts a public URL or Base64-encoded image (png, jpeg, webp, bmp, tiff, gif). Exactly one image; total pixels must be within [512x512, 6000x6000] and aspect ratio within [1/16, 16], max 30MB.

- **`size`** (`string`, _optional_):
  Output resolution tier. auto follows the input image's size (capped to the 1K-2K range). The base image keeps the input's aspect ratio; each layer keeps its own aspect ratio from the original.
  - Default: `"auto"`
  - Options: "auto", "1K", "1.5K", "2K"

- **`output_format`** (`string`, _optional_):
  File format of the base image only. Decomposed layers are always transparent png.
  - Default: `"jpeg"`
  - Options: "jpeg", "png"

- **`enable_sync_mode`** (`boolean`, _optional_):
  If set to true, the function will wait for the result to be generated and uploaded before returning the response. It allows you to get the result directly in the response. This property is only available through the API.
  - Default: `false`

- **`enable_base64_output`** (`boolean`, _optional_):
  If enabled, the output will be encoded into a BASE64 string instead of a URL. This property is only available through the API.
  - Default: `false`



**Required Parameters Example**:

```json
{
  "model": "bytedance/seedream-v5.0-flash/layer-decomposition",
  "image": ""
}
```


**Full Example**:

```json
{
  "model": "bytedance/seedream-v5.0-flash/layer-decomposition",
  "prompt": "",
  "image": "",
  "size": "auto",
  "output_format": "jpeg",
  "enable_sync_mode": false,
  "enable_base64_output": false
}
```


### Output Schema

The API returns the following output format:


- **`code`** (`integer`, _optional_):
  Status code of the call; 200 on success.

- **`message`** (`string`, _optional_):
  Status message of the call.

- **`data`** (`object`, _optional_):
  The prediction: id, model, status, outputs (base image first, then each layer in ascending z-order) and index-aligned layers metadata, plus urls and created_at. Poll GET /api/v1/model/prediction/{request_id} with data.id until status is completed or failed.



**Example Response**:

```json
{
  "code": 0,
  "message": "",
  "data": {}
}
```


## Usage Examples

### cURL

```bash
# Step 1: Start generation (async)
curl -X POST "https://api.atlascloud.ai/api/v1/model/generateImage" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "bytedance/seedream-v5.0-flash/layer-decomposition",
  "prompt": "",
  "image": "",
  "size": "auto",
  "output_format": "jpeg",
  "enable_sync_mode": false,
  "enable_base64_output": false
}'

# Response will contain: {"code": 200, "data": {"id": "prediction_id", "status": "processing"}}

# Step 2: Poll for result (replace {prediction_id} with the id returned above)
curl -X GET "https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY"

# Keep polling until status is "completed", "succeeded" or "failed"
# When completed, outputs will contain the generated content URL(s)
```

## Additional Resources

### Documentation

- [Model Playground](https://www.atlascloud.ai/models/bytedance/seedream-v5.0-flash/layer-decomposition)

1. Introduction

Seedream 5.0 Flash Layer Decomposition is the layer-separation variant of Seedream 5.0 Flash (Dola-Seedream-5.0-flash), ByteDance's high-speed, cost-optimized member of the Seedream 5.0 family added in September 2026 (upstream model version dola-seedream-5-0-flash-260915). It covers the API model identifier:

bytedance/seedream-v5.0-flash/layer-decomposition

Supply a single image and the model decomposes it into an editable stack: one base image plus up to 16 independent transparent layers, each returned with its stacking order, bounding-box coordinates, a machine-generated name, and a semantic description. It offers the same decomposition contract as Seedream 5.0 Pro Layer Decomposition (BytePlus ModelArk) on the faster Flash inference path. ByteDance positions Dola-Seedream-5.0-flash for instant creation: a high-speed, cost-optimized model covering text-to-image generation, single- and multi-reference image workflows, precision editing, and layer decomposition, aimed at latency-sensitive scenarios such as mobile assistant applications, UGC content creation, and bulk-scale content production.

2. Key Features & Innovations

Automatic Full-Image Decomposition: With no prompt, the model identifies every major element — text blocks, subjects, props, background — and splits each into its own layer, often ten or more in a single pass.
Three Targeting Modes: Omit the prompt for automatic decomposition; describe target elements in natural language; or pinpoint them with <bbox>x1 y1 x2 y2</bbox> tags using normalized [0, 1000] coordinates.
Transparent, Recomposable Assets: Every layer is a transparent PNG that keeps its own aspect ratio; the base image follows the input's aspect ratio. Stacking layers in z-order reconstructs the full picture.
Structured Layer Metadata: Each layer carries z_index, a bounding_box in absolute base-image pixels and normalized [0, 1000] coordinates, and a model-generated name and description, aligned one-to-one with the output image list.
Predictable Output Contract: At most 17 images per request (1 base + 16 layers), ordered base-first then ascending z-order, with all-or-nothing reliability — a request never returns a partial stack.

3. Model Architecture & Technical Details

Layer decomposition runs on the Seedream 5.0 family's multimodal transformer architecture: the planning stage analyses regional semantics of the input — segmenting elements, resolving stacking relationships, and scheduling both per-layer extraction and the inpainting of newly exposed base regions — before a diffusion-based decoder renders the stack in one pass.

Input is a single image (URL or Base64) in PNG or JPEG format, up to 30MB, with total pixels between 512×512 and 6000×6000 and aspect ratio between 1/16 and 16. Output resolution is selected by tier keyword — 1K, 1.5K, 2K, or auto (the default, which follows the input's own size within the 1K–2K envelope). The base image renders as JPEG or PNG per output_format; layers are always transparent PNG. Bounding boxes use the output base image's coordinate system with the origin at the top left.

4. Performance Highlights

Layer separation has no established public benchmark; observed behavior on typical inputs:

Scenario	Input	Result
Automatic full decomposition	16:9 storefront photo, `auto`	1 base + 11 named layers in ~75 seconds
Automatic full decomposition	wide product-shelf crop, `1K`	1 base + 12 named layers in ~75 seconds

Layer names and descriptions are specific enough to serve directly as layer-panel labels (for example "Window wall cup rack" or "Shop counter bar stools"), with bounding boxes that follow each element's extent.

5. Intended Use & Applications

Design Asset Extraction: Turn finished posters, banners, and product shots back into editable layered assets.
Localization & Copy Revision: Extract text elements as standalone layers, then translate, swap, or restyle them while the artwork beneath stays untouched.
Interactive Canvas Editors: Power click- and box-select editing — the <bbox> prompt syntax and per-layer bounding boxes map directly onto canvas coordinates.
E-Commerce Creative Repurposing: Isolate products, price tags, and promotional text, then recompose them across formats and marketplaces.
Bulk-Scale Asset Pipelines: Decompose large volumes of existing creatives into editable layers for UGC remix tools and automated content production.
Motion Graphics Preparation: Layered output with z-order and transparency feeds parallax, reveal, and micro-animation pipelines.

Seedream v5.0 Flash Layer Decomposition API by ByteDance

INPUT

OUTPUT

Parameters

Code Example

Install

Authentication

HTTP Headers

Submit a request

Submit a Request

Request Body

Response

Check Status

Polling Example

Status Values

Completed Response

Upload Files

Upload Example

Response

Input Schema

Example Request Body

Output Schema

Example Response

Atlas Cloud Skills

Supported Clients

Install

Setup API Key

Capabilities

MCP Server

Supported Clients

Install

Configuration

Available Tools

API Schema

LLM-Friendly Prompt Template

1. Introduction

2. Key Features & Innovations

3. Model Architecture & Technical Details

4. Performance Highlights

5. Intended Use & Applications

Explore Similar Models

Seedream v4.7 Edit Sequential

Seedream v4.7 Edit

Seedream v4.7 Sequential

Seedream v4.7 Text-to-Image

Seedream v5.0 Pro Layer Decomposition

Seedream v5.0 Flash Edit

Seedream v5.0 Flash Text-to-Image

Seedream v5.0 Pro Edit

Seedream v5.0 Pro Text-to-Image

Seed3D 2.0 Image-to-3D

Seedream v5.0 Lite Edit Sequential

Seedream v5.0 Lite Sequential

Seedream v5.0 Lite Edit

Seedream v5.0 Lite

Seedream v4.5

Seedream v4.5 Edit

One API for All Media AI.

INPUT

OUTPUT

Parameters

Code Example

Install

Authentication

HTTP Headers

Submit a request

Submit a Request

Request Body

Response

Check Status

Polling Example

Status Values

Completed Response

Upload Files

Upload Example

Response

Input Schema

Example Request Body

Output Schema

Example Response