
Separate elements from one PNG or JPEG image up to 30 MB. Choose 1K, 1.5K or 2K. A prompt is optional; when provided it must contain 4–5000 characters.
The final cost is rounded up to whole credits.

Example output
Seedream 5.0 Flash Layer Decomposition
Image to Layers

Example output
Seedream 5.0 Flash Layer Decomposition
Best for
Supports
Try this model
No results to display yet.
Model examples
Review examples to help decide whether the model fits your task.






Model strengths
Model strengths
Suggested uses
Tasks to consider
Model details
Provider
ByteDance
Category
Image
Capabilities
Image to Layers
Credit charges
Credit rate: 3.24 per image
Inputs
1 image reference
Prompt requirements
Optional prompt
Seedream 5.0 Flash Layer Decomposition FAQ
Separate elements from one PNG or JPEG image up to 30 MB. Choose 1K, 1.5K or 2K. A prompt is optional; when provided it must contain 4–5000 characters.
Credits are reserved before submission. Final billing uses confirmed consumption; any unused reserve is returned after settlement. If submission or consumption cannot be confirmed, the reserve remains held for reconciliation. Check credit history before sending again.
Compare alternatives
GPT Image 2.5 Flare supports image creation from a prompt and edits with image references, plus aspect ratio, resolution, and background settings.
Why consider it
Input option: Prompt + optional images
Seedream 5.0 Pro is Rivya's multi-reference image model for campaign frames, editorial scenes, and product compositions, with text-to-image, image-to-image, 1K or 2K output, and up to 10 reference images.
Why consider it
Input option: Prompt + optional images
Qwen Image 3.0 is Rivya's configurable Qwen image model for Chinese or English prompts, reference-led layouts, and reproducible visual tests with 1K or 2K output, prompt expansion, negative prompts, and seed control.
Why consider it
Input option: Prompt + optional images
Developer access
Seedream 5.0 Flash Layer Decomposition can be called through Public API v1. Modes that use reference media require uploads through the Files API. Check the model reference for supported inputs.