Search
AnimateDiff allows you to transform text and video prompts into dynamic videos with a wide range of motion types.
The prompt to use for generating the video. Be as descriptive as possible for best results.
The video to transform.
The negative prompt to use. Use it to address details that you don't want in the image.
The number of frames to generate. Value is ignored when video is provided.
The first N number of seconds of video to animate.
The number of inference steps to perform. 4-12 is recommended for turbo mode.
The strength of the input video in the final output.
The CFG scale is a measure of how close you want the model to stick to your prompt.
The same seed and prompt will output the same image every time.
Number of frames per second to extract from the video.
The motions to apply to the video.
The size of the video to generate.
Turbo mode is faster but may produce lower quality results.
The generated video from the model.