Skip to main content
The Wan 2.7 Image to Video node generates a video starting from a first-frame image. You can optionally provide a last-frame image to create a transition between the two, or provide an audio file to guide the video’s motion and timing. The node uses an AI model to animate the scene based on your text description.

Inputs

ParameterDescriptionData TypeRequiredRange
modelThe AI model to use for video generation.COMBOYes"wan2.7-i2v"
model.promptA text description of the elements and visual features you want in the video. Supports English and Chinese.STRINGYes-
model.negative_promptA text description of elements or features you want the model to avoid.STRINGYes-
model.resolutionThe resolution of the output video.COMBOYes"720P"
"1080P"
model.durationThe length of the generated video in seconds (default: 5).INTYes2 to 15
first_frameThe image to use as the first frame of the video. The output video’s aspect ratio is derived from this image.IMAGEYes-
last_frameAn optional image to use as the last frame. When provided, the model generates a video that transitions from the first frame to this last frame.IMAGENo-
audioAn optional audio file to drive the video generation, useful for lip-syncing or beat-matched motion. Duration must be between 2 and 30 seconds. If not provided, the model will generate matching background music or sound effects.AUDIONo-
seedA seed value to control the randomness of the generation (default: 0).INTYes0 to 2147483647
prompt_extendWhen enabled, the node will use AI assistance to enhance your text prompt (default: True). This is an advanced setting.BOOLEANYes-
watermarkWhen enabled, an AI-generated watermark will be added to the final video (default: False). This is an advanced setting.BOOLEANYes-
Note: The audio input has a duration constraint. If provided, the audio file must be between 2 and 30 seconds long.

Outputs

Output NameDescriptionData Type
outputThe generated video file.VIDEO
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

Source fingerprint (SHA-256): db5e4e26a407c71e2b79179d32cdad926db5af16461f2aa565eab544295df7ff