Wan 2.1 14b VACE

What you can set

• PromptRequired

  Text

• First image

  Upload an image

  URL of the first image.

• Images

  Upload one or more images

  URL of ref images to use while generating the video.

• Last image

  Upload an image

  URL of the last image.

• Mask image

  Upload an image

  URL of the mask image.

• Mask video

  Upload a video

  URL of the mask video.

• Video

  Upload a video

  The video for generating the output.

• Negative prompt

  Text

  The negative prompt for the generation.

• Context scale

  Number from 0 to 2

  Controls how close you want the model to stick to the reference context.

• Duration

  Number from 5 to 10

  The duration of the generated media in seconds.

• Flow shift

  Number from 0 to 30

  The shift value for the timestep schedule for flow matching.

• Guidance scale

  Number from 0 to 20

  The guidance scale to use for the generation.

• Steps

  Number from 1 to 40

  The number of inference steps to perform.

• Seed

  Number, repeats a result

  The random seed to use for the generation. -1 means a random seed will be used.

• Size

  Choice

  The size of the generated media in pixels (width*height).

  • 832*480 (default)
  • 480*832
  • 1280*720
  • 720*1280

• Task

  Choice

  Extract control information from the provided video to guide video generation.

  • depth (default)
  • pose
  • face
  • inpainting
  • none