Seedance 2.5 Video

Seedance 2.5 uses an asynchronous video task API: create a task first, then poll the retrieve endpoint for its status and final video URL. It uses the dedicated model name doubao-seedance-2.5, and its parameter ranges differ from Seedance 2.0.

Model selection

Use doubao-seedance-2.5 with both the unified and official-compatible video APIs.

For the Pro/Fast 2.0 models, see the separate Seedance 2.0 guide.

Endpoints

Capability	Method	Path
Create video task (official-compatible)	`POST`	`/api/v3/contents/generations/tasks`
Retrieve video task (official-compatible)	`GET`	`/api/v3/contents/generations/tasks/{task_id}`
Create video task	`POST`	`/v1/video/generations`
Retrieve video task	`GET`	`/v1/video/generations/{task_id}`
Create asynchronous image moderation task	`POST`	`/v1/images/moderations/tasks`
Retrieve asynchronous image moderation task	`GET`	`/v1/images/moderations/tasks/{task_id}`

Capabilities

Capability	Input
Text to video	Send only `text` items in `content`
Single-image video	Send one `image_url`; use `first_frame` or `reference_image`
First-and-last-frame video	Send two `image_url` items with `first_frame` and `last_frame`
Multiple reference images	Send multiple `reference_image` items and refer to them as “image 1”, “image 2”, and so on
Video reference	Use `video_url` with `role: "reference_video"`
Audio reference	Use `audio_url` with `role: "reference_audio"` together with text, an image, or a video
Image moderation	Moderate images first, then use the returned `asset://<asset ID>` in the generation request

Seedance 2.5 Parameters

Parameter	Required	Description
`model`	Yes	Use `doubao-seedance-2.5` with both the unified and official-compatible APIs
`content`	Yes	Array of text, image, video, and audio items; include at least text, an image, or a video
`duration`	No	`4` to `30` seconds; `-1` selects duration intelligently; usually defaults to `5`
`resolution`	No	Only `480p` and `720p` are supported
`ratio`	No	`adaptive`, `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, or `9:16`
`generate_audio`	No	Whether to generate synchronized audio
`watermark`	No	Whether to add a watermark
`seed`	No	Random seed; omit it to let the model select one
`return_last_frame`	No	When `true`, a successful result may include a final-frame image URL
`execution_expires_after`	No	Task execution timeout from `3600` to `259200` seconds
`tools`	No	Use `[ { "type": "web_search" } ]` to enable web search when available for the account
`safety_identifier`	No	Unique end-user identifier, up to 64 characters

For first-frame and first-and-last-frame tasks, use ratio: "adaptive". This preserves the input image ratio and keeps parameter validation consistent.

Output Pixel Sizes

Resolution	`16:9`	`4:3`	`1:1`	`3:4`	`9:16`	`21:9`
`480p`	854×480	752×560	640×640	560×752	480×854	992×432
`720p`	1280×720	1112×834	960×960	834×1112	720×1280	1470×630

With adaptive, the model selects an appropriate ratio from the input media and task. Read the final dimensions from the completed task.

`content` Items

content is an array whose items are selected by type. Image, video, and audio URLs must be nested inside the corresponding { "url": "..." } object; do not pass them as plain strings.

`type`	Required field	Optional `role`	Description
`text`	`text`	None	Text prompt; multiple text items are used in order
`image_url`	`image_url.url`	`reference_image`, `first_frame`, `last_frame`	Public image URL or an approved `asset://<asset ID>`
`video_url`	`video_url.url`	`reference_video`	Directly downloadable video URL; defaults to a reference video when `role` is omitted
`audio_url`	`audio_url.url`	`reference_audio`	Directly downloadable audio URL; defaults to reference audio when `role` is omitted

json

[
  { "type": "text", "text": "Describe the video to generate" },
  { "type": "image_url", "role": "reference_image", "image_url": { "url": "asset://reviewed-image-id" } },
  { "type": "video_url", "role": "reference_video", "video_url": { "url": "https://example.com/reference.mp4" } },
  { "type": "audio_url", "role": "reference_audio", "audio_url": { "url": "https://example.com/reference.mp3" } }
]

Common combinations include text only, text plus image, text plus video, text plus audio, and multimodal combinations of images, videos, and audio. Audio cannot be the only input; content must contain at least one text, image_url, or video_url item.

The platform accepts up to 30 reference images, 10 reference videos, and 10 reference audio files. Practical limits can also depend on file size, total media duration, and account permissions.

Text-to-Video

bash

curl -X POST https://cubicspaces.cloud/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "doubao-seedance-2.5",
    "content": [
      {
        "type": "text",
        "text": "A cinematic aerial shot of a futuristic city waking at sunrise, with a smooth forward camera movement"
      }
    ],
    "duration": 5,
    "resolution": "720p",
    "ratio": "16:9",
    "generate_audio": true,
    "watermark": false
  }'

A successful create request returns a public task ID:

json

{
  "id": "task_xxx",
  "task_id": "task_xxx",
  "object": "video",
  "model": "doubao-seedance-2.5",
  "status": "queued",
  "progress": 0,
  "created_at": 1780000000
}

Image Moderation

For people, products, or other library-managed reference images, submit public URLs to the image moderation API first. After approval, pass the returned items[].asset_url unchanged as content[].image_url.url.

All images used by the same generation request must be submitted together in one asynchronous moderation task. Do not split them across separate moderation batches.

Asynchronous Moderation

Use an asynchronous task for single images, multiple images, batches, and high concurrency:

bash

curl -X POST https://cubicspaces.cloud/v1/images/moderations/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "doubao-seedance-2.5",
    "images": [
      "https://example.com/person.png",
      "https://example.com/product.png"
    ],
    "client_request_id": "seedance-25-batch-001"
  }'

The create call immediately returns a task ID with status: "queued". One moderation task accepts up to 20 images.

json

{
  "code": "success",
  "message": "",
  "data": {
    "id": "amt_xxx",
    "model": "doubao-seedance-2.5",
    "status": "queued",
    "total": 2,
    "completed": 0,
    "approved": 0,
    "rejected": 0,
    "failed": 0
  }
}

Retrieve the moderation task:

bash

curl https://cubicspaces.cloud/v1/images/moderations/tasks/amt_xxx \
  -H "Authorization: Bearer YOUR_API_KEY"

Poll every 2 to 5 seconds. Processing statuses are queued and running; terminal statuses are succeeded, partial_succeeded, failed, and expired. Only image items with status: "approved" can use their asset_url in a generation request.

Image-to-Video

This example uses an asset_url returned by image moderation:

bash

curl -X POST https://cubicspaces.cloud/v1/video/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "doubao-seedance-2.5",
    "content": [
      {
        "type": "text",
        "text": "Preserve the appearance of the person in image 1. Show them playing with a small dog in a sunny meadow while keeping facial features consistent."
      },
      {
        "type": "image_url",
        "role": "reference_image",
        "image_url": {
          "url": "asset://reviewed-person-asset-id"
        }
      }
    ],
    "duration": 5,
    "resolution": "720p",
    "ratio": "16:9",
    "generate_audio": true
  }'

First and Last Frame

json

{
  "model": "doubao-seedance-2.5",
  "content": [
    {
      "type": "text",
      "text": "Transition smoothly from image 1 to image 2 while keeping the subject and camera motion coherent."
    },
    {
      "type": "image_url",
      "role": "first_frame",
      "image_url": { "url": "asset://first-frame-asset-id" }
    },
    {
      "type": "image_url",
      "role": "last_frame",
      "image_url": { "url": "asset://last-frame-asset-id" }
    }
  ],
  "duration": 8,
  "resolution": "720p",
  "ratio": "adaptive"
}

Video and Audio References

json

{
  "model": "doubao-seedance-2.5",
  "content": [
    {
      "type": "text",
      "text": "Use the camera motion from video 1 and the rhythm from audio 1 to create a new scene in the same style."
    },
    {
      "type": "video_url",
      "role": "reference_video",
      "video_url": { "url": "https://example.com/reference.mp4" }
    },
    {
      "type": "audio_url",
      "role": "reference_audio",
      "audio_url": { "url": "https://example.com/reference.wav" }
    }
  ],
  "duration": 10,
  "resolution": "720p",
  "ratio": "16:9"
}

Official-Compatible API

Existing official-style clients can use these paths directly:

http

POST /api/v3/contents/generations/tasks
GET /api/v3/contents/generations/tasks/{task_id}

Create a task:

bash

curl -X POST https://cubicspaces.cloud/api/v3/contents/generations/tasks \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "doubao-seedance-2.5",
    "content": [
      {
        "type": "text",
        "text": "A cinematic tracking shot of an orange cat running through a city street after the rain"
      }
    ],
    "duration": 5,
    "resolution": "720p",
    "ratio": "16:9",
    "generate_audio": true
  }'

The create response is an unwrapped official-compatible object:

json

{
  "id": "task_xxx",
  "model": "doubao-seedance-2.5",
  "status": "queued",
  "created_at": 1780000000
}

Retrieve the task:

bash

curl https://cubicspaces.cloud/api/v3/contents/generations/tasks/task_xxx \
  -H "Authorization: Bearer YOUR_API_KEY"

Example successful response:

json

{
  "id": "task_xxx",
  "model": "doubao-seedance-2.5",
  "status": "succeeded",
  "content": {
    "video_url": "https://example.com/generated-video.mp4"
  },
  "resolution": "720p",
  "ratio": "16:9",
  "duration": 5,
  "usage": {
    "completion_tokens": 108900,
    "total_tokens": 108900
  }
}

Retrieve a Unified Video Task

bash

curl https://cubicspaces.cloud/v1/video/generations/task_xxx \
  -H "Authorization: Bearer YOUR_API_KEY"

Use the public task_xxx returned by the create request and the same API key that created the task.

json

{
  "code": "success",
  "message": "",
  "data": {
    "task_id": "task_xxx",
    "status": "SUCCESS",
    "progress": "100%",
    "result_url": "https://example.com/generated-video.mp4",
    "properties": {
      "origin_model_name": "doubao-seedance-2.5"
    },
    "usage": {
      "prompt_tokens": 0,
      "completion_tokens": 108900,
      "total_tokens": 108900
    }
  }
}

data.usage.completion_tokens and data.usage.total_tokens are the final effective usage after the task completes. Before completion, data.usage may be omitted.

Status Values

API	Processing	Success	Failure or termination
Official-compatible	`queued`, `running`	`succeeded`	`failed`, `expired`, `cancelled`
Unified video	`NOT_START`, `SUBMITTED`, `QUEUED`, `IN_PROGRESS`	`SUCCESS`	`FAILURE`

Notes

Image, video, and audio URLs must be directly downloadable by the platform server. Do not use URLs that return login or HTML pages.
Images can use public URLs or asset://<asset ID> values returned by image moderation.
Submit all images for the same generation task in the same moderation batch.
Refer to media by order in the prompt, such as “image 1”, “video 1”, or “audio 1”. Do not place Asset IDs in the prompt.
content.video_url and data.result_url can expire. Download and retain the result promptly.
Model and extended-feature availability can vary by account permissions and current platform capabilities.

Seedance 2.5 Video ​

Endpoints ​

Capabilities ​

Seedance 2.5 Parameters ​

Output Pixel Sizes ​

content Items ​

Text-to-Video ​

Image Moderation ​

Asynchronous Moderation ​

Image-to-Video ​

First and Last Frame ​

Video and Audio References ​

Official-Compatible API ​

Retrieve a Unified Video Task ​

Status Values ​

Notes ​

Seedance 2.5 Video

Endpoints

Capabilities

Seedance 2.5 Parameters

Output Pixel Sizes

`content` Items

Text-to-Video

Image Moderation

Asynchronous Moderation

Image-to-Video

First and Last Frame

Video and Audio References

Official-Compatible API

Retrieve a Unified Video Task

Status Values

Notes