lucataco/wan-2.1-1.3b-vid2vid

Wan 2.1 1.3b Video to Video. Wan is a powerful visual generation model developed by Tongyi Lab of Alibaba Group

Capabilities

Negative PromptSeed

Community model (estimated from hardware time)

Name	Type	Description	Default	Constraints
`prompt`*	string	Text prompt describing what you want to generate or modify	`—`	—
`aspect_ratio`	string	Aspect ratio for the output video	`"832x480"`	832x480480x832
`cfg_scale`	number	Classifier free guidance scale (higher values strengthen prompt adherence)	`6`	min: 0, max: 20
`denoising_strength`	number	Strength of denoising when using video-to-video mode. Higher values change more content.	`0.7`	min: 0, max: 1
`frames_per_second`	integer	Number of frames per second in the output video	`16`	min: 5, max: 24
`input_video`	string(uri)	Input video for video-to-video generation	`—`	—
`negative_prompt`	string	Negative prompt to specify what to avoid in the generation	`"low quality, blurry, distorted, disfigured, text, watermark"`	—
`num_frames`	integer	Number of frames to generate in the output video	`81`	min: 1, max: 100
`num_inference_steps`	integer	Number of sampling steps (higher = better quality but slower)	`40`	min: 10, max: 50
`seed`	integer	Random seed for reproducible results (leave blank for random)	`—`	—
`tiled`	boolean	Whether to use tiled sampling for better quality on larger videos	`true`	—

promptrequiredstring

Text prompt describing what you want to generate or modify

aspect_ratiostring

Aspect ratio for the output video

Default: "832x480"

832x480480x832

cfg_scalenumber

Classifier free guidance scale (higher values strengthen prompt adherence)

Default: 6min: 0, max: 20

denoising_strengthnumber

Strength of denoising when using video-to-video mode. Higher values change more content.

Default: 0.7min: 0, max: 1

frames_per_secondinteger

Number of frames per second in the output video

Default: 16min: 5, max: 24

input_videostring

Input video for video-to-video generation

negative_promptstring

Negative prompt to specify what to avoid in the generation

Default: "low quality, blurry, distorted, disfigured, text, watermark"

num_framesinteger

Number of frames to generate in the output video

Default: 81min: 1, max: 100

num_inference_stepsinteger

Number of sampling steps (higher = better quality but slower)

Default: 40min: 10, max: 50

seedinteger

Random seed for reproducible results (leave blank for random)

tiledboolean

Whether to use tiled sampling for better quality on larger videos

Default: true

Version: 9349766527edUpdated: 7/25/20261.3K runs