> ## Documentation Index
> Fetch the complete documentation index at: https://doc.starrise.tech/llms.txt
> Use this file to discover all available pages before exploring further.

# Seedance 2.0 Ultra

> Call ByteDance's Seedance Ultra video generation model via Starrise AI API — supports multimodal references, video editing, extension, audio generation, and web search enhancement.

Seedance 2.0 Ultra is ByteDance's latest video generation model, available through the Starrise AI API. It supports text-to-video, image-to-video, multimodal reference input, video editing, video extension, and synchronized audio generation.

## Key Capabilities

| Feature                               | Description                                                                                      |
| ------------------------------------- | ------------------------------------------------------------------------------------------------ |
| Text-to-video                         | Generate video from a text description                                                           |
| Image-to-video (first frame)          | Use one image as the first frame                                                                 |
| Image-to-video (first and last frame) | Use two images as the first and last frames respectively                                         |
| Multimodal reference                  | Combine images, video, and audio as references (1–9 images, up to 3 videos, up to 3 audio clips) |
| Video editing                         | Modify elements in an existing video using a reference image                                     |
| Video extension                       | Extend and concatenate reference videos                                                          |
| Audio generation                      | Automatically generate synchronized voice, sound effects, and background music                   |
| Web search enhancement                | Enhance generation with real-time internet content (text-to-video only)                          |
| Return last frame                     | Retrieve the last frame of the generated video                                                   |

## Output Specifications

| Property     | Value                                     |
| ------------ | ----------------------------------------- |
| Resolution   | 720p, 1080p, 2k                           |
| Aspect ratio | 16:9, 4:3, 1:1, 3:4, 9:16, 21:9, adaptive |
| Duration     | 4–15 seconds                              |
| Format       | mp4                                       |

## Workflow

```
1. POST /v1/video/generations  →  task_id
2. Poll GET /v1/video/generations/{task_id}  →  status
3. When status = "succeeded"  →  download video URL (valid for 24 hours)
```

## Examples

### Text-to-Video

<CodeGroup>
  ```bash cURL theme={null}
  curl https://ai.alad.com/v1/video/generations \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "seedance-2.0-ultra",
      "content": [
        {
          "type": "text",
          "text": "A cat playing piano in a sunlit room, cinematic lighting"
        }
      ],
      "generate_audio": true,
      "ratio": "16:9",
      "duration": 8
    }'
  ```

  ```python Python theme={null}
  import requests
  import time

  headers = {"Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json"}

  # Step 1: Create task
  resp = requests.post(
      "https://ai.alad.com/v1/video/generations",
      headers=headers,
      json={
          "model": "seedance-2.0-ultra",
          "content": [
              {"type": "text", "text": "A cat playing piano in a sunlit room, cinematic lighting"}
          ],
          "generate_audio": True,
          "ratio": "16:9",
          "duration": 8
      }
  )
  task_id = resp.json()["task_id"]

  # Step 2: Poll for results
  while True:
      result = requests.get(
          f"https://ai.alad.com/v1/video/generations/{task_id}",
          headers=headers
      ).json()
      status = result["data"]["status"]
      if status == "SUCCESS":
          print("Video URL:", result["data"]["data"]["content"]["video_url"])
          break
      elif status == "FAILED":
          print("Failed:", result["data"].get("fail_reason"))
          break
      time.sleep(5)
  ```
</CodeGroup>

### Multimodal Reference (Image + Video + Audio)

In the prompt you can reference `content` array items using `@image1`, `@video1`, `@audio1`, numbered by media type starting from 1.

> **Important:** Assets must be passed in strict order: text, image\_url, video\_url, audio\_url. Do not reorder them, as this may cause errors; when including multiple assets, also ensure no other asset types are mixed in.

```bash cURL theme={null}
curl https://ai.alad.com/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-ultra",
    "content": [
      {
        "type": "text",
        "text": "The character from @image1 dances in the scene from @video1, with @audio1 as background music, festive New Year atmosphere"
      },
      {
        "type": "image_url",
        "image_url": {"url": "https://example.com/character.jpg"},
        "role": "reference_image"
      },
      {
        "type": "video_url",
        "video_url": {"url": "https://example.com/clip.mp4"},
        "role": "reference_video"
      },
      {
        "type": "audio_url",
        "audio_url": {"url": "https://example.com/bgm.mp3"},
        "role": "reference_audio"
      }
    ],
    "generate_audio": true,
    "ratio": "16:9",
    "duration": 11
  }'
```

### Video Editing

```bash cURL theme={null}
curl https://ai.alad.com/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-ultra",
    "content": [
      {
        "type": "text",
        "text": "Replace the water in @video1 with the perfume from @image1, keep the same camera motion"
      },
      {
        "type": "image_url",
        "image_url": {"url": "https://example.com/perfume.jpg"},
        "role": "reference_image"
      },
      {
        "type": "video_url",
        "video_url": {"url": "https://example.com/original.mp4"},
        "role": "reference_video"
      }
    ],
    "generate_audio": true,
    "ratio": "16:9",
    "duration": 5
  }'
```

### Web Search Enhancement (Text-to-Video Only)

```bash cURL theme={null}
curl https://ai.alad.com/v1/video/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-ultra",
    "content": [
      {
        "type": "text",
        "text": "Macro shot of vivid flower petals on a tree, gradually zooming in"
      }
    ],
    "generate_audio": true,
    "ratio": "16:9",
    "duration": 11,
    "tools": [{"type": "web_search"}]
  }'
```

## Request Parameters

| Parameter                 | Type    | Required       | Description                                                                          |
| ------------------------- | ------- | -------------- | ------------------------------------------------------------------------------------ |
| `model`                   | string  | Yes            | `seedance-2.0-ultra`                                                                 |
| `content`                 | array   | Yes            | Input content array (text, image\_url, video\_url, audio\_url)                       |
| `content[].type`          | string  | Yes            | `text`, `image_url`, `video_url`, or `audio_url`                                     |
| `content[].text`          | string  | For text type  | Text prompt (recommended ≤500 Chinese chars or ≤1000 English words)                  |
| `content[].image_url.url` | string  | For image type | Image URL, Base64, or `asset://<ID>`                                                 |
| `content[].video_url.url` | string  | For video type | Video URL or `asset://<ID>` (mp4/mov, ≤50 MB, 2–15 seconds)                          |
| `content[].audio_url.url` | string  | For audio type | Audio URL, Base64, or `asset://<ID>` (wav/mp3, ≤15 MB)                               |
| `content[].role`          | string  | Conditionally  | `first_frame`, `last_frame`, `reference_image`, `reference_video`, `reference_audio` |
| `generate_audio`          | boolean | No             | Generate synchronized audio, default `true`                                          |
| `resolution`              | string  | **Yes**        | `720p`, `1080p`, or `2k`. Required for Ultra.                                        |
| `ratio`                   | string  | No             | `16:9`, `4:3`, `1:1`, `3:4`, `9:16`, `21:9`, `adaptive`, default `adaptive`          |
| `duration`                | integer | No             | 4–15 seconds, default `5`                                                            |
| `tools`                   | array   | No             | `[{"type": "web_search"}]` enables web search (text-to-video only)                   |
| `watermark`               | boolean | No             | Whether to add a watermark, default `false`                                          |

### Input Modes

| Mode                                  | Content items                         | `role` values                                           |
| ------------------------------------- | ------------------------------------- | ------------------------------------------------------- |
| Text-to-video                         | 1× text                               | —                                                       |
| Image-to-video (first frame)          | text (optional) + 1× image\_url       | `first_frame` or omitted                                |
| Image-to-video (first and last frame) | text (optional) + 2× image\_url       | `first_frame` + `last_frame`                            |
| Multimodal reference                  | text (optional) + images/videos/audio | `reference_image`, `reference_video`, `reference_audio` |
| Video editing                         | text + image\_url + video\_url        | `reference_image` + `reference_video`                   |
| Video extension                       | text + video\_url(s)                  | `reference_video`                                       |

> **Note:** First frame, first-and-last-frame, and multimodal reference are three mutually exclusive modes that cannot be combined.

### Referencing Assets in Prompts

In the text prompt you can use `@<type><N>` placeholders to reference media items in the `content` array, **numbered in order of appearance within each media type**:

| Placeholder             | Referenced item              |
| ----------------------- | ---------------------------- |
| `@image1`, `@image2`, … | 1st, 2nd, … `image_url` item |
| `@video1`, `@video2`, … | 1st, 2nd, … `video_url` item |
| `@audio1`, `@audio2`, … | 1st, 2nd, … `audio_url` item |

Example: `"The character from @image1 walks through the scene in @video1, with @audio1 as background music"`.

> **Note:** Assets must be passed in strict order: text, image\_url, video\_url, audio\_url. Do not reorder them, as this may cause errors; when including multiple assets, also ensure no other asset types are mixed in.

### Resolution and Pixel Values by Aspect Ratio

| Resolution | 16:9      | 4:3       | 1:1       | 3:4       | 9:16      | 21:9      |
| ---------- | --------- | --------- | --------- | --------- | --------- | --------- |
| 720p       | 1280x720  | 1112x834  | 960x960   | 834x1112  | 720x1280  | 1470x630  |
| 1080p      | 1920x1080 | 1664x1248 | 1440x1440 | 1248x1664 | 1080x1920 | 2208x944  |
| 2k         | 2560x1440 | 2224x1668 | 1920x1920 | 1668x2224 | 1440x2560 | 2944x1260 |

<Card title="API Reference" icon="code" href="/en/api-reference/model-api/bytedance/seedance-2-0-ultra">
  View the interactive API documentation for Seedance 2.0 Ultra.
</Card>