## Basic model info

Model name: bytedance/seedance-2.0-fast
Model description: A faster variant of Seedance 2.0 for quicker video generation with multimodal inputs and native audio.


## Model inputs

- prompt (required): Text prompt for video generation (string)
- image (optional): Input image for image-to-video generation (first frame). Cannot be combined with reference images. (string)
- last_frame_image (optional): Input image for last frame generation. Only works if a first frame image is also provided. Cannot be combined with reference images. (string)
- reference_images (optional): Reference images (up to 9) for character consistency, style guidance, and scene composition. Cannot be used together with first/last frame images. You can reference them in your prompt as [Image1], [Image2], etc. (array)
- reference_videos (optional): Reference videos (up to 3, total duration max 15s) for motion transfer, style reference, and editing. Reference them in your prompt as [Video1], [Video2], etc. (array)
- reference_audios (optional): Reference audio files (up to 3, total duration max 15s) for audio-driven generation and lip-sync. Requires at least one reference image or video. Reference them in your prompt as [Audio1], [Audio2], etc. (array)
- duration (optional): Video duration in seconds. Set to -1 for intelligent duration (model picks the best length). (integer)
- resolution (optional): Video resolution. (string)
- aspect_ratio (optional): Video aspect ratio. Set to 'adaptive' to let the model choose the best ratio based on inputs. (string)
- generate_audio (optional): Generate synchronized audio with the video, including dialogue (use double quotes in prompt), sound effects, and background music. (boolean)
- seed (optional): Random seed. Set for reproducible generation. (integer)


## Model output schema

{
  "type": "string",
  "title": "Output",
  "format": "uri"
}

If the input or output schema includes a format of URI, it is referring to a file.


## Example inputs and outputs

Use these example outputs to better understand the types of inputs the model accepts, and the types of outputs the model returns:

### Example (https://replicate.com/p/0mw6h7wpgnrmt0cxdsr98t0fbm)

#### Input

```json
{
  "seed": 99,
  "prompt": "A cozy cabin in a snowy forest at night, warm light glowing from the windows, gentle snowfall, camera slowly pushing in through the trees",
  "duration": 7,
  "resolution": "720p",
  "aspect_ratio": "16:9",
  "generate_audio": true
}
```

#### Output

```json
"https://replicate.delivery/xezq/iZvjneezp8uI2kef2HwWq3cmIr8Nve7YFYqJm67a49M2KrIzC/tmp8xcrlyhq.mp4"
```


### Example (https://replicate.com/p/4v66kq831xrmw0cxdsp836kka8)

#### Input

```json
{
  "seed": 42,
  "prompt": "A golden retriever puppy chasing butterflies through a sunlit meadow, soft bokeh background, cinematic camera slowly tracking the puppy",
  "duration": 5,
  "resolution": "720p",
  "aspect_ratio": "16:9",
  "generate_audio": true
}
```

#### Output

```json
"https://replicate.delivery/xezq/cBahBCuXrlYHGhSgkftFqHXnBngtkI1MJawXIfmkoeN5oKysA/tmp20l933am.mp4"
```


### Example (https://replicate.com/p/k3whvchva1rmy0cxdsptapqfjm)

#### Input

```json
{
  "seed": 123,
  "prompt": "A sushi chef carefully preparing an intricate sushi roll, close-up overhead shot, steam rising, warm restaurant lighting",
  "duration": 5,
  "resolution": "720p",
  "aspect_ratio": "16:9",
  "generate_audio": true
}
```

#### Output

```json
"https://replicate.delivery/xezq/6htcFBOadLLfWyr884rLXPwDLbCy8lh3UedyQkJCPwDMWFZWA/tmpbub0c7bf.mp4"
```


### Example (https://replicate.com/p/mmhsxbqhksrmt0cxdsqbbcfwdc)

#### Input

```json
{
  "seed": 77,
  "prompt": "Aerial drone shot of a turquoise ocean wave curling and crashing on white sand beach, golden hour light, slow motion",
  "duration": 8,
  "resolution": "720p",
  "aspect_ratio": "21:9",
  "generate_audio": true
}
```

#### Output

```json
"https://replicate.delivery/xezq/eiuIHOr6A3SGCaLuEBG2JfwcWLUHlRz8k8qfxipda9ejfqIzC/tmpwunnu9mx.mp4"
```


### Example (https://replicate.com/p/51t3y50895rmt0cxdss9t0rmg4)

#### Input

```json
{
  "seed": 256,
  "prompt": "A woman in a flowing red dress walking along the edge of a cliff overlooking the sea, wind blowing her hair and dress, dramatic wide angle, golden sunset",
  "duration": 5,
  "resolution": "720p",
  "aspect_ratio": "9:16",
  "generate_audio": true
}
```

#### Output

```json
"https://replicate.delivery/xezq/3ln0eT6HtyRjNi3e4Frrk6ft6OcGbkK00AxREweNejoNYrIzC/tmplllxppif.mp4"
```


## Model readme

> # Seedance 2.0 Fast
> 
> A faster variant of [Seedance 2.0](/bytedance/seedance-2.0) that trades some quality for speed. It shares the same multimodal architecture and capabilities — text-to-video, image-to-video, multimodal reference, video editing, and video extension — but generates results more quickly.
> 
> Use Seedance 2.0 Fast when iteration speed matters more than maximum quality, like early prototyping or high-volume content pipelines. Switch to [Seedance 2.0](/bytedance/seedance-2.0) for final renders where quality is the priority.
> 
> ## Capabilities
> 
> Seedance 2.0 Fast supports the same input modes as Seedance 2.0:
> 
> - **Text to video** — describe a scene and get a video with synchronized audio.
> - **Image to video** — animate a still image as the first frame, optionally with a last frame.
> - **Multimodal reference** — combine up to 9 images, 3 video clips, and 3 audio files. Reference them in your prompt as [Image1], [Video1], [Audio1], etc.
> - **Video editing** — provide a reference video and describe changes to make.
> - **Video extension** — continue an existing video with a text description of what comes next.
> - **Native audio** — generates synchronized dialogue, sound effects, and background music alongside the video. For dialogue, put speech in double quotes in your prompt.
> - **Intelligent duration** — set duration to -1 and the model picks the best length.
> - **Adaptive aspect ratio** — set to "adaptive" and the model chooses the best ratio for your inputs.
> 
> ## Supported resolutions
> 
> | Resolution | 16:9 | 4:3 | 1:1 | 3:4 | 9:16 | 21:9 |
> |---|---|---|---|---|---|---|
> | 480p | 864×496 | 752×560 | 640×640 | 560×752 | 496×864 | 992×432 |
> | 720p | 1280×720 | 1112×834 | 960×960 | 834×1112 | 720×1280 | 1470×630 |
> 
> ## Tips
> 
> - Be specific in prompts — describe camera movements, lighting, mood, and actions.
> - For dialogue, use double quotes: `She turned and said: "Follow me."`
> - Label reference inputs in your prompt: "The cat from [Image1] does the dance from [Video1]."
> - Start with 5-second clips while prototyping, then increase duration for final versions.
> 
> ## Learn more
> 
> For full details on Seedance 2.0's architecture and capabilities, see the [official page](https://seed.bytedance.com/en/seedance2_0).
> 
> For the higher-quality variant, see [Seedance 2.0](/bytedance/seedance-2.0).
> 
> You can try this model on the [Replicate Playground](https://replicate.com/playground).