ListenHubDocs
API ReferenceImage Generation

Seedream

Seedream 5.0 Pro limits and precise region editing driven by pixel coordinates and color codes in the prompt.

Seedream 5.0 Pro is ByteDance's image model on ListenHub, and the only one in the catalogue that supports precise editing — changing a specified region of an existing image instead of regenerating the whole thing. It runs on the shared image endpoints. For the endpoints, request schema, async flow, and error codes, see the Image Generation overview. This page covers only what is specific to Seedream.

Model

ModelDisplay nameBest for
seedream-5-0-proSeedream 5.0 ProEditing a specific region of a supplied image

Takes provider: "bytedance".

Limits

LimitValue
Image sizes1K, 2K — 4K returns 400
Aspect ratiosAll 14, including the extreme four
Default aspect ratio1:1 from the request schema
Reference imagesUp to 10
qualityIgnored

Seedream is validated on size, not on ratio: every ratio the schema accepts is in bounds, including the four extreme ones that Nano Banana allows only on Flash. Output pixels are computed from the size budget and the aspectRatio you send — which the schema defaults to 1:1, so set it explicitly to match your source image. The one hard limit is size: 4K is rejected outright with code: 29003 rather than silently downgraded to 2K.

Precise editing

There is no separate edit endpoint and no mask or region parameter. Editing uses the same generation endpoint, with the source image in referenceImages and the "what to change, and how" written into prompt. The model reads absolute pixel coordinates (origin at the top-left corner of the image) and industry color codes directly from the prompt text.

{
  "provider": "bytedance",
  "model": "seedream-5-0-pro",
  "prompt": "Treating the top-left corner as the coordinate origin, change the content inside top-left:(376,363) bottom-right:(701,638) to green, and leave everything else unchanged",
  "referenceImages": [
    {
      "fileData": {
        "fileUri": "https://assets.listenhub.app/your-source-image.png",
        "mimeType": "image/png"
      }
    }
  ],
  "imageConfig": { "imageSize": "2K", "aspectRatio": "1:1" }
}

The server passes the prompt through verbatim and does not parse coordinates. If your product offers visual editing gestures (box select, lasso, arrows), translate them into coordinate strings in your own frontend — or bake the annotations into the reference image — before writing them into prompt. Keep aspectRatio matching the source image's real ratio; a mismatch can make the model reconstruct the whole image instead of editing one region.

Pricing

Seedream is priced at a fixed credit cost per image size (pricing.mode is fixed) and never sets requiresSubscription. It has a free-quota resource (seedream-5-0-pro) spendable at 1K and 2K — which is every size it supports. Call POST /v1/images/generation/estimate-credits for the exact figure.

Example

curl -X POST "https://api.marswave.ai/openapi/v1/images/generation/async" \
  -H "Authorization: Bearer $LISTENHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "provider": "bytedance",
    "model": "seedream-5-0-pro",
    "prompt": "Treating the top-left corner as the coordinate origin, repaint the jacket inside top-left:(120,240) bottom-right:(520,880) in #1F4D3A, keep the face, hair and background unchanged",
    "referenceImages": [
      { "fileData": { "fileUri": "https://example.com/portrait.png", "mimeType": "image/png" } }
    ],
    "imageConfig": { "imageSize": "2K", "aspectRatio": "3:4" }
  }'

See the Image Generation overview for the full parameter list and the reference-image request shapes.

On this page