Go to OpenArt main site
Upload workflow

Vid2Vid Style Transfer with IPA & Hotshot XL

5.0

2 reviews
312
38.2K
10.2K
52
Description

Found that the "Strong Style Transfer" of IPAdapter performs exceptionally well in Vid2Vid.

Just update your IPAdapter and have fun~!


Checkpoint I used:

Any turbo or lightning model will be good, like Dreamshaper XL Turbo or lightning, Juggernaut XL lightning etc. (Remember to check the required samplers and lower your CFG)


HotshotXL download:

Download: https://huggingface.co/Kosinkadink/HotShot-XL-MotionModels/tree/main

Context_length: 8 is good


And I recommend you take a look at Inner_Reflections_AI 's article, which provides very detailed explanations: https://civitai.com/articles/2601



A Lower Memory use vae:

Download: https://civitai.com/models/140686?modelVersionId=155933



Discussion

(No comments yet)

Loading...

Author

3
20.1K
719
86.9K

No reviews yet

  • - latest (8 months ago)

Primitive Nodes (1)

Anything Everywhere3 (1)

Custom Nodes (36)

AnimateDiff Evolved

  • - ADE_AnimateDiffLoaderGen1 (1)

  • - ADE_StandardUniformContextOptions (1)

ComfyUI

  • - ControlNetApplyAdvanced (3)

  • - VAELoader (1)

  • - CLIPTextEncode (2)

  • - VAEEncode (2)

  • - KSampler (2)

  • - VAEDecode (2)

  • - LoraLoader (1)

  • - UpscaleModelLoader (1)

  • - CheckpointLoaderSimple (1)

  • - ImageScaleBy (1)

  • - ImageUpscaleWithModel (1)

  • - ImageScale (2)

  • - LoadImage (1)

  • - AIO_Preprocessor (3)

  • - IPAdapterUnifiedLoader (1)

  • - IPAdapterAdvanced (1)

  • - ControlNetLoaderAdvanced (3)

  • - VHS_VideoCombine (4)

  • - VHS_LoadVideoPath (1)

  • - ColorMatch (1)

Checkpoints (1)

dreamshaperXL_v21TurboDPMSDE.safetensors

LoRAs (1)

OIL_ON_CANVAS_v3.safetensors