floyo logo
Powered by
ThinkDiffusion
⚡️Nano Banana 2 ⚡️ just landed. Start creating now.
floyo logo
Powered by
ThinkDiffusion
⚡️Nano Banana 2 ⚡️ just landed. Start creating now.

Wan2.1 Fun Control and Flux for V2V Restyle

Create a new video by restyling an existing video with a reference image.

2.2k

Generates in about 14 mins 35 secs

Nodes & Models

GetNode
WanVideoTeaCache
WanVideoSLG
UNETLoader
flux1-dev-fp8.safetensors
DualCLIPLoader
clip_l.safetensors
t5xxl_fp8_e4m3fn.safetensors
VAELoader
Note
Label (rgthree)
PrimitiveInt
Fast Groups Muter (rgthree)
WanVideoLoraSelect
Wan/wan_i2v_720_l1v3w4llp4p3r_e50_with_trigger.safetensors
WanVideoModelLoader
Wan2.1-Fun-Control-14B_fp8_e4m3fn.safetensors
WanVideoVAELoader
wan_2.1_vae.safetensors
LoadWanVideoT5TextEncoder
umt5_xxl_fp16.safetensors
Fast Groups Bypasser (rgthree)
LoadWanVideoClipTextEncoder
open-clip-xlm-roberta-large-vit-huge-14_fp16.safetensors
LoadImage
AddLabel
CLIPTextEncodeFlux
PreviewImage
SetNode
ImageResizeKJ
WanVideoTextEncode
ImageBlend
WanVideoEncode
EmptyLatentImage
WanVideoControlEmbeds
WanVideoClipVisionEncode
VAEDecode
WanVideoImageToVideoEncode
SaveImage
WanVideoSampler
WanVideoDecode
ImageConcatMulti
LoadFluxControlNet
flux-depth-controlnet-v3.safetensors
ApplyFluxControlNet
XlabsSampler
Seed Everywhere
VHS_LoadVideo
VHS_SelectImages
VHS_VideoInfo
VHS_VideoCombine
VHS_LoadVideo
VHS_SelectImages
VHS_VideoInfo
VHS_VideoCombine
easy positive
easy float
MiDaS-DepthMapPreprocessor
Scribble_XDoG_Preprocessor
DWPreprocessor
AIO_Preprocessor
MiDaS-DepthMapPreprocessor
Scribble_XDoG_Preprocessor
DWPreprocessor
AIO_Preprocessor
DWPreprocessor
ImpactMakeImageBatch
CM_NearestSDXLResolution
ImageResize+
ImageResize+
RIFE VFI
rife47.pth
FinalFrameSelector
FinalFrameSelector

Create a new video by restyling an existing video with a reference image.

Restyle lets you apply the visual characteristics of one image (like a painting style) to another image or video. Unlike basic filters, AI-powered style transfer analyzes both the content and artistic elements, creating results that look like they were created in that style from scratch.

With Wan 2.1, you can transform ordinary videos into ones that appear hand-painted, animated, or rendered in specific artistic styles while maintaining the original movement and content.

Key Inputs

Load Reference Video: Use any Mp4 that you would like to transform

Image reference: Use any JPG or PNG showing your subject clearly in the style you'd like

Set Video Prompt: as descriptive a prompt as possible.

Width & height: Optimal resolution settings are noted, 1280x720 by default.

Frame load cap: how many desired frames in the output video (for Wan, it must be a multiple of 4, plus 1. for example, 65, or 101)

Skip first frames: Start the AI process somewhere in further into the video's timeline

Select every nth frame: Only process every other Nth frame in the generated video

Read more

N