Skip to content

inference.partner.seedream-5-0.pro.img2img.v1

View markdown

The inference.partner.seedream-5-0.pro.img2img.v1 job uses Dola Seedream 5.0 Pro.

{
"type": "inference.partner.seedream-5-0.pro.img2img.v1",
"config": {
"images": [
"reference.png"
],
"prompt": "A fox in a snowy forest"
}
}

Supports resolution (1K/1.5K/2K or custom dimensions), watermark, and image output. Image-to-image accepts up to 10 reference images. layer_decomposition: true requires exactly one image and returns the base image, PNG layers, and layers.json describing the names, positions, and stacking order of the layers. resolution: "auto" is available for layers. Transparent output requires a single reference image with an alpha channel.

{
"type": "object",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"type",
"config"
],
"properties": {
"type": {
"type": "string"
},
"config": {
"type": "object",
"additionalProperties": false,
"properties": {
"image": {
"type": "string",
"format": "filename"
},
"images": {
"type": "array",
"minItems": 1,
"maxItems": 10,
"items": {
"type": "string",
"format": "filename"
}
},
"prompt": {
"description": "Text prompt; optional for automatic layer decomposition.",
"type": "string"
},
"resolution": {
"description": "Output tier or widthxheight. auto is supported only for layer decomposition.",
"type": "string",
"default": "2K",
"pattern": "^(1K|1\\.5K|2K|auto|[1-9][0-9]*x[1-9][0-9]*)$"
},
"watermark": {
"type": "boolean",
"default": false
},
"prompt_optimization": {
"type": "string",
"default": "standard",
"enum": [
"standard",
"fast"
]
},
"layer_decomposition": {
"type": "boolean",
"default": false
},
"background": {
"type": "string",
"default": "opaque",
"enum": [
"opaque",
"transparent"
]
}
},
"allOf": [
{
"not": {
"required": [
"image",
"images"
]
}
}
]
}
}
}