Skip to content

Get Result

Request

Result queries can use the response_url from the create-task response, or build the request path with BASE_URL, endpoint, and request_id.

Use response_url
Query result by request_id

curl -sS "${RESPONSE_URL}" \
  "${COMMON_HEADERS[@]}"

curl -sS "${BASE_URL}/v1/queue/${ENDPOINT}/requests/${REQUEST_ID}/response" \
  "${COMMON_HEADERS[@]}"

Response

Incomplete response

If the task has not reached a terminal state, the result endpoint returns HTTP 202 with a body that matches status queries:

202

{
  "status": "IN_PROGRESS",
  "request_id": "mmsu_01hxyz...",
  "response_url": "https://cloud.seaart.ai/model/v1/queue/nano_banana_2/requests/mmsu_01hxyz.../response"
}

Completed response

After a task succeeds, the result endpoint returns a stable structure: request_id, outputs, and metadata. Every downloadable or displayable result is placed in the outputs array. Multiple result URLs are returned as multiple records in order.

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "image",
      "url": "https://example.com/image.png",
      "content_type": "image/png",
      "file_name": "image.png",
      "width": 1024,
      "height": 1024
    }
  ],
  "metadata": {
    "prompt": "a cinematic photo of a cat astronaut"
  }
}

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "image",
      "url": "https://example.com/image-1.png",
      "content_type": "image/png",
      "width": 1024,
      "height": 1024
    },
    {
      "type": "image",
      "url": "https://example.com/image-2.png",
      "content_type": "image/png",
      "width": 1024,
      "height": 1024
    }
  ],
  "metadata": {
    "prompt": "generate two product variations"
  }
}

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "video",
      "url": "https://example.com/video.mp4",
      "content_type": "video/mp4",
      "file_name": "video.mp4",
      "file_size": 123456,
      "duration": 5
    }
  ],
  "metadata": {
    "prompt": "camera slowly pushes in"
  }
}

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "audio",
      "url": "https://example.com/audio.mp3",
      "content_type": "audio/mpeg",
      "file_name": "audio.mp3",
      "duration": 10.5
    }
  ],
  "metadata": {
    "text": "你好，欢迎使用 SeaCloud。"
  }
}

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "model3d",
      "url": "https://example.com/model.glb",
      "content_type": "model/gltf-binary",
      "file_name": "model.glb",
      "file_size": 123456
    }
  ],
  "metadata": {
    "prompt": "a small stylized robot, game-ready asset"
  }
}

{
  "request_id": "mmsu_01hxyz...",
  "outputs": [
    {
      "type": "text",
      "text": "[Verse]\n晚风吹过海岸线\n我们把故事唱一遍"
    }
  ],
  "metadata": {
    "prompt": "写一首关于夏夜海边和朋友重逢的中文流行歌词"
  }
}

{
  "status": "COMPLETED",
  "request_id": "mmsu_01hxyz...",
  "response_url": "https://cloud.seaart.ai/model/v1/queue/nano_banana_2/requests/mmsu_01hxyz.../response",
  "error": "Provider rejected request",
  "error_type": "REQUEST_INVALID"
}

`outputs` field

Field	Type	Description
`type`	string	Result type. Common values are `image`, `video`, `audio`, `model3d`, `text`, and `file`
`url`	string	Downloadable or accessible result URL. Text results may omit this field
`text`	string	Text result content, such as generated lyrics
`content_type`	string	Resource MIME type, such as `image/png` or `video/mp4`
`file_name`	string	File name, returned when the platform can infer it
`file_size`	number	File size in bytes, returned when the platform can obtain it
`width` / `height`	number	Image width and height, returned when the platform can obtain them
`duration`	number	Audio or video duration in seconds, returned when the platform can obtain it
`metadata`	object	Additional information for a single output, omitted when there is no extra information