Available now

Wan VACE 14B Depth: depth-guided video

By Alibaba

Wan VACE 14B Depth reads the depth of a source video and renders a new scene from your prompt with the same shapes, layout and camera motion.

Model ID
Maker
Alibaba

Input

Fields marked * are required. Review the price before generating.

The text prompt to guide video generation.

Example prompt. Edit it or write your own.

URL to the source video file. Required for depth task.

Uploading a file needs an account. Create account

MP4, WebM · up to 100 MB

The source reference image. If provided, the model will use this image as reference.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

If provided, the model will use this frame as a reference.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

If provided, the model will use this frame as a reference.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

Additional settings

Must be between 81 to 241 (inclusive).

Must be between 5 to 30.

Random seed for reproducibility. If None, a random seed is chosen.

Number of inference steps for sampling. Higher values give better quality but take longer.

Guidance scale for classifier-free guidance. Higher values encourage the model to generate images closely related to the text prompt.

Shift parameter for video generation.

Whether to enable prompt expansion.

Whether to preprocess the input video.

Number of frames to interpolate between the original frames. A value of 0 means no interpolation.

The model to use for frame interpolation. Options are 'rife' or 'film'.

If true, also return a ZIP file containing all generated frames.

Price$0.1268 / second

You sign up first. Your settings are kept.

About this model

Overview

Upload a video and describe the new scene. The depth of every frame guides the result, so objects keep their place and distance while their look changes. Reference images and a first or last frame can steer the look further; output is 480p, 580p or 720p.

What you can build

  • A filmed room turned into a different interior style
  • Live footage restyled as a fantasy or sci-fi scene
  • Rough previs or block-out footage rendered with a finished look

Prompting tips

  • Describe surfaces, light and style; the depth already fixes where everything is.
  • Use footage with clear foreground and background so the depth reads well.

What to check

Compare foreground silhouettes, depth boundaries and camera movement throughout the clip.

Inputs

  • Prompt (required): The text prompt to guide video generation.
  • Negative Prompt (optional): Negative prompt for video generation.
  • Number of Frames (optional): Number of frames to generate. Must be between 81 to 241 (inclusive).
  • Frames per Second (optional): Frames per second of the generated video. Must be between 5 to 30.
  • Seed (optional): Random seed for reproducibility. If None, a random seed is chosen.
  • Resolution (optional): Resolution of the generated video. Options: 480p, 580p, 720p.
  • Aspect Ratio (optional): Aspect ratio of the generated video. Options: auto, 16:9, 1:1, 9:16.
  • Number of Inference Steps (optional): Number of inference steps for sampling. Higher values give better quality but take longer. From 2 to 50.
  • Guidance Scale (optional): Guidance scale for classifier-free guidance. Higher values encourage the model to generate images closely related to the text prompt. From 1 to 10.
  • Sampler (optional): Sampler to use for video generation. Options: unipc, dpm++, euler.
  • Shift (optional): Shift parameter for video generation. From 1 to 15.
  • Video (required): URL to the source video file. Required for depth task.
  • Reference Images (optional): URLs to source reference image. If provided, the model will use this image as reference.
  • First Frame (optional): URL to the first frame of the video. If provided, the model will use this frame as a reference.
  • Last Frame (optional): URL to the last frame of the video. If provided, the model will use this frame as a reference.
  • Enable Prompt Expansion (optional): Whether to enable prompt expansion.
  • Preprocess (optional): Whether to preprocess the input video.
  • Acceleration (optional): Acceleration to use for inference. Options are 'none' or 'regular'. Accelerated inference will very slightly affect output, but will be significantly faster. Options: none, low, regular.
  • Video Quality (optional): The quality of the generated video. Options: low, medium, high, maximum.
  • Video Write Mode (optional): The write mode of the generated video. Options: fast, balanced, small.
  • Number of Interpolated Frames (optional): Number of frames to interpolate between the original frames. A value of 0 means no interpolation. From 0 to 5.
  • Temporal Downsample Factor (optional): Temporal downsample factor for the video. This is an integer value that determines how many frames to skip in the video. A value of 0 means no downsampling. For each downsample factor, one upsample factor will automatically be applied. From 0 to 5.
  • Enable Auto Downsample (optional): If true, the model will automatically temporally downsample the video to an appropriate frame length for the model, then will interpolate it back to the original frame length.
  • Auto Downsample Min FPS (optional): The minimum frames per second to downsample the video to. This is used to help determine the auto downsample factor to try and find the lowest detail-preserving downsample factor. The default value is appropriate for most videos, if you are using a video with very fast motion, you may need to increase this value. If your video has a very low amount of motion, you could decrease this value to allow for higher downsampling and thus longer sequences. From 1 to 60.
  • Interpolator Model (optional): The model to use for frame interpolation. Options are 'rife' or 'film'. Options: rife, film.
  • Transparency Mode (optional): The transparency mode to apply to the first and last frames. This controls how the transparent areas of the first and last frames are filled. Options: Content aware, white, black.
  • Return Frames Zip (optional): If true, also return a ZIP file containing all generated frames.

Output and limits

  • Resolutions: 480p, 580p, 720p.
  • Aspect ratios: auto, 16:9, 1:1, 9:16.
  • Readable text inside the video, especially small text, is unreliable.
  • Scenes with several interacting characters can lose consistency.
  • Longer clips can show more motion drift or scene changes.
  • Low-quality inputs cause glitches: heavy compression, collages, screenshots and text overlays.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • per second: $0.1268 / second
  • The total for your settings is rounded up to the next cent.