About Seedream 5.5 Layers
Seedream 5.5 Layers splits a single image into a clean base image and separate transparent PNG layers, one per element: subjects, objects, decorations and blocks of text. Each layer keeps its position on the base, so a poster, ad or product shot can be rebuilt and rearranged in a design tool.
Leave the prompt blank to split out every main element, or name the elements to pull apart. With text recognition on, each text layer also reports its wording, a similar font family and its color, ready for retyping or translation. A split returns up to 99 layers, with a ZIP of the base, every layer and a manifest of their names, order and positions.
See what this model can do
A handful of real prompts and their outputs.
Ready to build with Seedream 5.5 Layers?
Try Seedream 5.5 Layers in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/images/edit \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "bytedance-seedream-5-5-layers",
"input_image": "https://hub.oxen.ai/api/repos/ox/Oxen-AI-Assets/file/main/images/ox_zoom_out_1926_1076.png",
"max_layers": 10,
"text_recognition": true,
"output_format": "png",
"watermark": false,
"layer_decomposition": true
}'Request parameters
Every field Seedream 5.5 Layers accepts in the request body. See the API reference for response formats.
https://hub.oxen.ai/api/ai/images/edit| Field | Type | Required | Default | Description |
|---|---|---|---|---|
input_image | stringuri | Yes | — | The image to split into layers. |
prompt | string | No | — | Optional. Name the elements to split out, such as "the cyclist, the scooter and the bakery sign". Leave blank to split out every main element. |
max_layers | integer | No | 10 | The most layers to return, not counting the base image. Higher values split the image more finely, and each layer is billed as an image. Range: 1 – 99 |
text_recognition | boolean | No | true | Read each text layer's wording, a similar font family and its color. |
output_format | string | No | "png" | Format of the base image.pngjpeg |
watermark | boolean | No | false | Whether to add an 'AI generated' watermark. |
layer_decomposition | boolean | No | true |