Type | Workflows |
Stats | 191 0 |
Reviews | (9) |
Published | Mar 31, 2025 |
Base Model | |
Hash | AutoV2 7F90B38BAD |
ποΈ ComfyUI Image-to-Video Workflow - WAN 2.1 Wrapper (Kiko WAN v3)
This is a high-performance, multi-pass Image-to-Video workflow for ComfyUI, powered by the WAN 2.1 Wrapper, with advanced optimizations like torch.compile
and Sage Attention for faster and smarter frame generation. I tried to expose all the settings that Kijai exposes that I can understand π, This is not the fastest workflow you will find on here, but it is one I use to make 20 secons videos.
Crafted with β€οΈ on Arch Linux BTW, using an RTX 4090 and 128 GB of RAMβthis setup is tuned for heavy-duty inference and silky-smooth video generation.
π Features
π§ WAN 2.1 Wrapper for cinematic image-to-video transformations
π Two-pass generation: initial + refinement/extension
π Optional Slow Motion + Frame Interpolation (RIFE, FILM, etc.)
π§½ Sharpening and Upscaling (e.g., RealESRGAN, SwinIR)
π οΈ Includes torch.compile for faster inference
π Integrates Sage Attention for improved attention efficiency
π Customizable prompts, seed, duration, and aspect ratio logic
π Final loop polish with "Extend Last Frame"
βοΈ System Specs
OS: Arch Linux (rolling release)
GPU: NVIDIA RTX 4090 (24GB VRAM)
RAM: 128 GB DDR5
Python: 3.12.9 via
pyenv
ComfyUI: Latest build from GitHub
torch: 2.x with
torch.compile
enabledSage Attention: Enabled via patched attention mechanism
π οΈ Workflow Overview
πΉ Input & Resize
Drop an image and optionally resize to fit WAN 2.1's expected input.
πΉ WAN 2.1 Wrapper Core
Uses
torch.compile
for speed boostEnhanced with Sage Attention (set via the custom node or environment)
πΉ Pass 1: Generate + Optional Slow Motion
Frame-by-frame synthesis
Add slow motion via interpolation node (RIFE or FILM)
πΉ Pass 2: Extend + Merge
Extends the motion, ensures smoother transitions
Combines motion with refined prompt guidance
πΉ Final Polish
Sharpening and Upscaling
Final interpolation if needed
Loop-ready output by extending the last frame
π§ͺ Performance Tips
Tune torch compile for you system, they are all different, my setting might not work for you.
For Sage Attention:
Use the node
Running on lower-end GPUs? Disable upscaling and reduce frame count.
π§° Requirements
ComfyUI
WAN 2.1 Wrapper Node
Optional:
RIFE
,FILM
, orDAIN
for interpolationRealESRGAN
/SwinIR
for upscalingSage Attention
patch or node
βΆοΈ How to Use
Load the
kiko-wan-v3.json
file into ComfyUI.Drop your image into the input node.
Customize prompts, duration, and frame count.
Click
Queue Prompt
to generate.Your video will be rendered in the output folder.
π Files
kiko-wan-v3.json
β Exported workflow (coming soon)kiko-wan-v3.png
β Workflow diagram
π§ Inspirations & Credits
Real-ESRGAN, RIFE, FILM, Sage Attention contributors
Arch Linux + NVIDIA ecosystem for elite workstation performance π
π‘ Future Plans
Add batch image-to-video mode
Audio?
βοΈ Custom Nodes Used in kiko-wan-wrapper-v3.json
Anything Everywhere: https://github.com/chrisgoringe/cg-use-everywhere
Display Any (rgthree): https://github.com/rgthree/rgthree-comfy
Fast Bypasser (rgthree): https://github.com/rgthree/rgthree-comfy
Fast Groups Bypasser (rgthree): https://github.com/rgthree/rgthree-comfy
GetImageRangeFromBatch: https://github.com/kijai/ComfyUI-KJNodes
GetImageSize+: https://github.com/cubiq/ComfyUI_essentials
Image Filter: https://github.com/chrisgoringe/cg-image-filter
ImageBatchMulti: https://github.com/kijai/ComfyUI-KJNodes
ImageFromBatch+: https://github.com/cubiq/ComfyUI_essentials
ImageListToImageBatch: https://github.com/ltdrdata/ComfyUI-Impact-Pack
ImageResizeKJ: https://github.com/kijai/ComfyUI-KJNodes
LoadWanVideoClipTextEncoder: https://github.com/kijai/ComfyUI-WanVideoWrapper/
LoadWanVideoT5TextEncoder: https://github.com/kijai/ComfyUI-WanVideoWrapper/
MarkdownNote: NOT FOUND
PlaySound|pysssss: https://github.com/pythongosssss/ComfyUI-Custom-Scripts
ProjectFilePathNode: https://github.com/MushroomFleet/DJZ-Nodes
RIFE VFI: https://github.com/Fannovel16/ComfyUI-Frame-Interpolation
ReActorRestoreFace: https://github.com/Gourieff/ComfyUI-ReActor
Seed Generator: https://github.com/giriss/comfy-image-saver
SimpleMath+: https://github.com/cubiq/ComfyUI_essentials
Text Input [Dream]: https://github.com/alt-key-project/comfyui-dream-project
VHS_VideoCombine: https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite
WanVideoBlockSwap: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoDecode: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoEnhanceAVideo: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoFlowEdit: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoImageClipEncode: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoLoopArgs: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoLoraBlockEdit: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoLoraSelect: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoModelLoader: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoSLG: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoSampler: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoTeaCache: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoTextEncode: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoTorchCompileSettings: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoVAELoader: https://github.com/kijai/ComfyUI-WanVideoWrapper/
WanVideoVRAMManagement: https://github.com/kijai/ComfyUI-WanVideoWrapper/