Text to Image API

{
  "allOf": [
    {
      "type": "object",
      "properties": {
        "name": {
          "type": "string",
          "description": "Name of the text to image task (optional, auto-generated if not provided)",
          "example": "My Generated Image"
        },
        "prompt": {
          "type": "string",
          "description": "Text prompt for image generation",
          "example": "A beautiful sunset over mountains"
        },
        "creation_mode": {
          "type": "string",
          "enum": [
            "text-to-image",
            "image-edit"
          ],
          "description": "Optional compatibility field for history and mode reporting; callers may omit it. The server ignores any supplied value and records `image-edit` when `input_images` contains a valid HTTP(S) image URL, otherwise `text-to-image`. It is not forwarded to the model provider."
        },
        "width": {
          "type": "number",
          "description": "Image width",
          "default": 1024,
          "example": 1024
        },
        "height": {
          "type": "number",
          "description": "Image height",
          "default": 1024,
          "example": 1024
        },
        "model_type": {
          "type": "string",
          "description": "Model type to use for generation. Use `a2e` for automatic model selection.",
          "enum": [
            "a2e",
            "zimage",
            "seedream"
          ],
          "default": "a2e",
          "example": "a2e"
        },
        "model_version": {
          "type": "string",
          "description": "Seedream model version. Required to select Seedream 4.5, 5.0, or 5.0 Pro when model_type is seedream.",
          "enum": [
            "4.5",
            "5.0",
            "5.0-pro"
          ],
          "default": "4.5",
          "example": "5.0-pro"
        },
        "input_images": {
          "type": "array",
          "description": "Reference images for the auto-selected or Seedream model. Auto selection supports at most 2 images; Seedream 5.0 Pro supports at most 10.",
          "items": {
            "type": "string"
          },
          "maxItems": 10,
          "example": [
            "https://example.com/image1.jpg",
            "https://example.com/image2.jpg"
          ]
        },
        "skip_face_enhance": {
          "type": "boolean",
          "description": "Whether to disable face similarity enhancement for auto-selected image editing. Defaults to false.",
          "default": false,
          "example": false
        },
        "aspect_ratio": {
          "type": "string",
          "description": "Output aspect ratio for the Seedream family. Rejected with 400 when the value is outside this enum.\nSeedream 5.0 Pro resolves the size from this ratio first and only falls back to `width`/`height` when it is absent, and the backend defaults it to `9:16` for Seedream requests that omit it.\n",
          "enum": [
            "1:1",
            "4:3",
            "3:4",
            "16:9",
            "9:16",
            "2:3",
            "3:2",
            "21:9"
          ],
          "example": "1:1"
        },
        "resolution": {
          "type": "string",
          "description": "Output resolution tier. Rejected with 400 when the value is outside this enum.\nThe billed tier is the higher of this value and the tier inferred from the short side of `width`/`height`, so a large `width`/`height` cannot be billed as a lower tier. `a2e` and `zimage` only have `1K`, `1080P`, and `2K`; `3K`/`4K` are treated as `2K` for them. For Seedream 5.0 Pro the output size comes from `aspect_ratio` plus this tier, so `width`/`height` never raise an explicitly requested tier there.\nEffective tiers per Seedream version: `5.0-pro` distinguishes only `1K` and `2K` and treats every other value as `2K`; `4.5` and `5.0` raise `1K`/`1080P` to `2K` because Seedream requires at least 3,686,400 pixels, and pass `3K`/`4K` through (`4.5` serves the `4K` tier, `5.0` serves `3K`).\n",
          "enum": [
            "1K",
            "1080P",
            "2K",
            "3K",
            "4K"
          ],
          "example": "2K"
        },
        "max_images": {
          "type": "integer",
          "description": "Maximum number of images to generate (creates multiple tasks internally)",
          "minimum": 1,
          "maximum": 8,
          "example": 1
        },
        "minor_suspected_skip": {
          "type": "boolean",
          "default": false,
          "description": "Set to true when retrying after error code 1004 to confirm and bypass the suspected-minor soft block."
        }
      },
      "required": [
        "prompt"
      ]
    },
    {
      "$ref": "#/components/schemas/WebhookInput"
    }
  ]
}