Go to OpenArt main site
Upload workflow

Ultimate Video Generation V3.2 (AnimateDiff, SVD, IP-Adapter, ControlNet, Upscaler ...)

5.0

0 reviews
43
9.4K
2.3K
1
Description

You are able to run only part of the workflow instead of always running the entire workflow.


# How to use

1. Install custom node from You will need custom node:

   https://github.com/komojini/komojini-comfyui-nodes

2. Configure Settings on "Inputs & Settings" group

3. Configure models and some other parameters on "Set AnimatedDiff Context" group.


There are different colors of workflows for Text-to-Video or Video-to-Video


You should set the input video by using the GetNode, currently there are:

- INPUT_IMAGES

- ANIMATED_DIFF_IMAGES

- ANIMATED_DIFF_FACEID_IMAGES

- HI_RES_FIX_IMAGES

- FACE_SWAP_RESTORE_IMAGES

- MERGED_IMAGES

- UPSCALED_IMAGES


# Features


You can set the sequence of your workflow by setting these input videos(images) as you need. For example:

- INPUT_IMAGES => ANIMATED_DIFF_IMAGES => HI_RES_FIX_IMAGES => UPSCALE_IMAGES

- INPUT_IMAGES => UPSCALE_IMAGES

- INPUT_IMAGES => FACE_SWAP_RESTORE_IMAGES => UPSCALE_IMAGES

- etc.


You can disable workflows that you are not using by setting "bypass" the VideoCombine Node on the right end.


If there are any issues or comments please share your thoughts!

Discussion

(No comments yet)

Loading...

Author

10
14.2K
223
52.1K

No reviews yet

  • - latest (9 months ago)

  • - v20240119-011238

  • - v20240119-005313

  • - v20240118-172318

  • - v20240118-155340

  • - v20240118-074707

Primitive Nodes (23)

Context (rgthree) (1)

Context Big (rgthree) (5)

Display Int (rgthree) (3)

IPAdapterApplyFaceID (1)

InsightFaceLoader (1)

Integer (1)

KSampler Config (rgthree) (1)

Note (7)

Seed (rgthree) (1)

SetNode (2)

Custom Nodes (151)

AnimateDiff Evolved

  • - CheckpointLoaderSimpleWithNoiseSelect (1)

  • - ADE_AnimateDiffLoaderWithContext (3)

  • - ADE_AnimateDiffModelSettings_Release (1)

  • - ADE_AnimateDiffUniformContextOptions (1)

  • - CR Apply LoRA Stack (1)

  • - CR LoRA Stack (1)

ComfyUI

  • - UpscaleModelLoader (1)

  • - VAELoader (1)

  • - VAEEncode (2)

  • - CLIPTextEncode (1)

  • - FreeU_V2 (1)

  • - LoraLoader (1)

  • - CLIPVisionLoader (1)

  • - VAEDecode (4)

  • - ImageUpscaleWithModel (1)

  • - LoadImage (2)

  • - VideoLinearCFGGuidance (1)

  • - SVD_img2vid_Conditioning (1)

  • - ImageOnlyCheckpointLoader (1)

  • - KSampler (1)

  • - ImageResize+ (1)

  • - GetImageSize+ (1)

  • - IterativeLatentUpscale (1)

  • - PixelKSampleUpscalerProvider (1)

  • - IPAdapterModelLoader (1)

  • - AV_ControlNetEfficientStacker (5)

  • - String (4)

  • - VHS_VideoCombine (6)

  • - Apply ControlNet Stack (2)

  • - FaceRestoreModelLoader (1)

  • - FaceRestoreCFWithModel (1)

  • - BatchPromptScheduleLatentInput (1)

  • - FloatConstant (1)

  • - KSamplerCacheable (2)

  • - Getter (52)

  • - Setter (34)

  • - ImageMerger (1)

  • - UltimateVideoLoader (1)

  • - FlowBuilder (6)

  • - ReActorFaceSwap (1)

Checkpoints (2)

SVD/svd_xt.safetensors

dreamshaper_8.safetensors

LoRAs (1)

ip-adapter-faceid-plusv2_sd15_lora.safetensors