June 10, Update: stmfnet
Just to save folks a headache the file is available for download as a model now.
June 06, Update: V4.0
Just a minor improvement, using a double interpolation of SFTMVFI results in higher fidelity of dynamic or fast motion scenes. Video is manually loaded to reduce time when generating high quantities of test runs. Now you can select which videos you want to upscale, will copy the name of the original input video.
Included the context framing node from Kijai's wrapper. Usually leave this off will exponentially increase generation times if on; however, can let you generate much longer videos beyond what your vram will allow you.
Interpolation requires higher VRAM if scaling if greater than 1.0. Changed the scaling factor to float for more control.
Updated nodes if there is any issues feel free to let me know.
https://drive.google.com/file/d/1s5JJdt5X69AO2E2uuaes17aPwlWIQagG/view
Place it in the folder: Custom_nodes\ComfyUI-Frame-Interpolation\ckpts\stmfnet
April 14, Update: V3.3
Updated Nodes.
March 27, Update: V3.2
Small changes to the Torch Compile and Tea Cache for more dynamic outputs using my Lora's.
Workflow is suitable for a 24GB card using the 14B models and a 12GB card using the 1.3B model.
Updates:
***Error within the naming of outputs corrected.
Uploaded a very simple and straightforward Wan I2V workflow. The workflow includes an upscaler with frame interpolation to bring your videos to a cool and smooth 60FPS. No BS just a simple addition. Original WanI2V is based on the workflow example of the amazing work from Kijai linked here: https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/main/example_workflows.
Additionally, I've left the annotations within.
Prev Updates:
Compatible with 12GB memory cards. Tested using a 3060 RTX 12GB using this workflow can comfortably go up to 85 frames. 97 Frames is pushing the absolute limits will take about an hour to generate.
Due to demand this workflow is a small improvement to the Super Simple HV2V for those who don't do postprocessing outside of Comfyui.
Designed for use with Hunyuan Video and trained Lora models, it can upscale videos to true 60 FPS. It combines detailed notes on denoising strength, step optimization, aspect ratio adjustments, frame interpolation, and upscaling to create high-quality, smooth videos.
Order matters:
Generate → Upscale → Interpolate. Upscaling first preserves detail for interpolation.
Adjust parameters based on your Lora’s behavior and hardware limits.
Description
Latest update for wrappers and nodes
FAQ
Comments (22)
With v3.3 since you switched out the ImageClip for ImageEncode, will it still auto-adjust the resolution?
it does, you can set the limits for you generation and if will auto adjust the image resolution within the set bounds of width and height.
Getting ''NoneType' object has no attribute 'lower'' on the WanVideo Model Loader?
Strange, did you install the wanvideo nodes from Kijais Got or using the Comfyui Manager?
@SingularUnity From the Comfyui manager, but I managed to fix it by disabling the Lora loader for now
@BubbleButtPrincesses strange, I cloned the repo directly to the custom nodes folder. Instructions are on Repo page.
@SingularUnity I'll keep playing with it
Correct me if I’m wrong but the triton error I’m getting for the Wan Video Wrapper, isn’t Triton only compatable on Linux? So this workflow would be dead for anyone using windows right? Do you have any idea as what to use in its place then?
You can remove the triton node and it will work but it'll be slower to gen. However, you can install Triton on windows fairly easily now. I've been using it for months before the simple windows fork was developed, it was possible then its possible now.
https://github.com/woct0rdho/triton-windows
@SingularUnity Thanks man, when I get back home I will give it a shot 👍
@User5000f24 Triton is also now working on windows, and is fairly easy to install
Well, nodes from this workflow bricked my ComfyUI to the point of no return, had to do a clean install and re-install/download everything.
Did you figure out why this happened?
@SingularUnity Unfortunately I did not, I simply reinstalled everything from scratch and never touched this workflow again.
@droegan that's so strange, I'm sorry that happened to you. I periodically backup my python_embedded folder to avoid those kinds of inconveniences. It's happened more than a few times, usually when trying to implement a new feature.
@SingularUnity Not your fault, all workflows are good to share, ComfyUI is picky sometimes is all I guess. :)
Sadly, I run out of memory and Comfy crashes. 4070 with 12GB VRAM here. Any tips?
adjust the resolution and change the model to the 1.3B parameters. With your VRAM youd be hard pressed to do anything on the bigger models unfortunately.
Use multiGPU GGUF nodes! native wf! I am also on a RTX 4070 12GB, 64 GB RAM. I am using for ex all the Q8 gguf wan checkpoints 17+GB. Gen time 832x832 3 secs is 472/its, 5 secs 945/its. Q4,5,6 is not significant faster from my testings so I stay with the quality!
3060 with 16gb here, get a bigger vram
Do you also have a t2v workflow for this one?
I want to use 1 or more character loras.
Hey just saw this, adapting it to a t2v should be easy enough. Simply use an empty latent instead of an image and change the model type from i2v to t2v