Windows Installation Guide for UniAnimate and Animate-X in ComfyUI

This guide walks you through the step-by-step installation process for UniAnimate and Animate-X nodes in ComfyUI -Windows. Follow these instructions carefully to set everything up properly.

 

System Requirements (Minimum)

Operating System: Windows (Windows 10/11 recommended)

RAM: 16GB

VRAM: 12GB Nvidia Graphics Card

 

ComfyUI Environment (Main Requirements): These are the libraries I have used to test the installation without issues. These requirements are not strict. What matters is that you use libraries that are compatible with each other)

ComfyUI GPU version

python>=3.9

Git

cuda 11.8 or 12.1

torch:

pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2

xFormers = 0.0.20

Or

pytorch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1)

xformers==0.0.27

Or

If you use another version of pytorch, make sure you use compatible versions of pytorch-cuda, torchvision, torchaudio and xformers. See the links below to know the compatible versions.

https://pytorch.org/get-started/previous-versions/

https://github.com/facebookresearch/xformers

https://anaconda.org/xformers/xformers/files

 

The custom nodes

The following are the custom nodes that will be installed.

Align & Generate poses for UniAnimate

Repose image with UniAnimate

Animate image with UniAnimate

Animate image with UniAnimate_Long

Repose image with Animate_X

Animate image with Animate_X

Animate image with Animate_X_Long

 

Installation Method 1 (Use the ComfyUI  Manager)

This installation via the ComfyUI manager assumes that your ComfyUI torch version is 2.3.1 and therefore attempts to install other required libraries that are compatible with this version of torch. If you have a different torch version and you want to stick to it, then it’s better you install the nodes with the second method.

The nodes can be found in this github repository: https://github.com/Isi-dev/ComfyUI-UniAnimate-W You can visit it to know more about the project and it can also assist you with the installation. But we will proceed with the easiest way to install the nodes. You can use another installation of ComfyUI to avoid requirement conflicts with other custom nodes.

Step 1:

Launch your ComfyUI application

Step 2:

Click on the "Manager" button in the main menu and select the "Custom Nodes Manager" button

Step 3:

In the search bar, type " UniAnimate Nodes for ComfyUI"

Step 4:

Click the "Install" button next to the node.

Sometimes the installation might fail due to an unstable network, so keep trying until the installation succeeds.

Step 5:

Click the "Restart" button in the Custom Nodes Manager

Step 6:

Head to your ‘ComfyUI\custom_nodes\ComfyUI-UniAnimate-W-main\’ folder and drag any of the .json files (e.g. uniAnimateReposeImg.json) into your ComfyUI interface.

Step 7:

If any of the nodes appear red, then check your ComfyUI terminal for any error. You can use ChatGPT to troubleshoot or visit the issues section of the above github link to see if a similar error has been resolved. If not, you can raise an issue or make a comment down below. You can also check the comment sections of the following videos for replies to some of the issues raised by others:

https://youtu.be/vR8EHoAQziI

https://youtu.be/NFnhELV4bG0

Step 8:

If you do not get a red node, then visit this huggingface repository: https://huggingface.co/camenduru/unianimate/tree/main and download the five checkpoints you find there.

Also visit this huggingface repository to download the animate-x checkpoint: https://huggingface.co/Shuaishuai0219/Animate-X/tree/main

You only need to download the animate-x_ckpt.pth checkpoint. All other checkpoints are the same as those downloaded from the huggingface repository first mentioned.

Place the checkpoints in your ' ComfyUI \custom_nodes\ComfyUI-UniAnimate-W-main\checkpoints' folder.

Close your ComfyUI terminal and launch it again. You should now be able to use the nodes.


 

Installation Method 2 (Use your Command Prompt)

Step 1:

Open a terminal (CMD) and navigate to the custom_nodes directory in your ComfyUI folder:

cd ComfyUI\Custom_nodes

Step 2:

Clone the ComfyUI-UniAnimate-W repository:

git clone https://github.com/Isi-dev/ComfyUI-UniAnimate-W.git

This step might fail if your network is unstable, so keep trying until it succeeds.

Step 3:

If using a virtual environment e.g. conda, you can now activate it:

conda activate <your_env_name>

But if using a system or python-embedded environment, you can proceed to the next step.

Step 4:

Navigate to the ComfyUI-UniAnimate-W-main folder in your terminal:

cd ComfyUI-UniAnimate-W-main

Step 5:

Check your torch version with your terminal.

=>In a conda virtual environment:

conda list torch

=>In a python-embedded environment:

..\..\..\python_embeded\python.exe -c "import torch; print(torch.__version__)"

=>If using your system environment (not recommended):

python -c "import torch; print(torch.__version__)"

I will now focus on the conda and python-embedded environments. The difference between the python-embedded environment and your system environment are the prefix: ..\..\..\python_embeded\python.exe & python. Every other thing is the same.

If your torch version is 2.0.1, then you should see something like: 2.0.1+cu118 which indicates that this pytorch version supports cuda version 11.8. If you see something like 2.0.1+cpu, then you will need to reinstall torch by following this guide: https://pytorch.org/get-started/previous-versions/

If you want to install  torch version 2.5.0 (I have not used this version so I don’t know if it is backwards compatible with the versions I have tried)

=>In a conda virtual environment:

conda uninstall torch torchvision torchaudio pytorch-cuda 

# CUDA 11.8

conda install pytorch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0  pytorch-cuda=11.8 -c pytorch -c nvidia

# CUDA 12.1

conda install pytorch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 pytorch-cuda=12.1 -c pytorch -c nvidia

# CUDA 12.4

conda install pytorch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 pytorch-cuda=12.4 -c pytorch -c nvidia

Verify the installation:

python -c "import torch; print(torch.__version__); print(torch.cuda.is_available())"

 

=>In a python-embedded environment:

..\..\..\python_embedded\python.exe -m pip uninstall torch torchvision torchaudio

..\..\..\python_embedded\python.exe -m pip uninstall pytorch-cuda

# CUDA 11.8

..\..\..\python_embeded\python.exe  -m pip install torch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 --index-url https://download.pytorch.org/whl/cu118

# CUDA 12.1

..\..\..\python_embeded\python.exe  -m pip install torch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 --index-url https://download.pytorch.org/whl/cu121

# CUDA 12.4

..\..\..\python_embeded\python.exe  -m pip install torch==2.5.0 torchvision==0.20.0 torchaudio==2.5.0 --index-url https://download.pytorch.org/whl/cu124

Verify the installation:

..\..\..\python_embeded\python.exe -c "import torch; print(torch.__version__); print(torch.cuda.is_available())"

You should see e.g. 2.5.0+cul118 true.

If you see 2.5.0+cul118 false, verify if your system has a CUDA-compatible GPU:

nvidia-smi

If this command fails, then install the appropriate NVIDIA drivers from NVIDIA's website

Step 6:

If your pytorch version is 2.5.0, then install xformers version 0.0.28.post2.

=> In a conda virtual environment:

pip install xformers==0.0.28.post2

=> In a python-embedded environment:

..\..\..\python_embedded\python.exe -m pip install xformers==0.0.28.post2

Make sure you install the xformers version compatible with your torch version. Shown below are the pytorch versions with their compatible xformers  versions based on these sites (https://github.com/facebookresearch/xformers/releases?page=1  & https://anaconda.org/xformers/xformers/files )and on people’s experience:

torch==2.0.1 xformers ==0.0.20

torch==2.2.2 xformers == 0.0.25.post1

torch==2.3.0 xformers == 0.0.26.post1

torch==2.3.1 xformers == 0.0.27

torch==2.4.0 xformers == 0.0.27.post2

torch==2.5.0 xformers == 0.0.28.post2

torch==2.5.1 xformers == 0.0.28.post3

You can analyze the second site to know about other compatible versions.

Step 7:

You may need to install the following libraries depending on your environment. Some people skipped this step and were still able to use the nodes, while some installed some of the libraries based on the errors in their terminals:

numpy

opencv-python

pytorch_lightning

lightning_utilities

lightning_fabric

torchmetrics

einops

onnxruntime

open-clip-torch

fairscale

easydict

imageio

matplotlib

args

Step 8:

Launch your comfyUI application and open your ‘ComfyUI\custom_nodes\ComfyUI-UniAnimate-W-main\’ folder. Drag any of the .json files (e.g. uniAnimateReposeImg.json) in the folder into your ComfyUI interface.

Step 9:

If any of the nodes appear red, then check your ComfyUI terminal for any error. You can use ChatGPT to troubleshoot or visit the issues page of this github repository: https://github.com/Isi-dev/ComfyUI-UniAnimate-W to see if a similar error has been resolved. If not, you can raise an issue or make a comment down below. You can also check the comment sections of the following videos for replies to some of the issues raised by others:

https://youtu.be/vR8EHoAQziI

https://youtu.be/NFnhELV4bG0

Step 10:

If you do not get a red node, then visit this huggingface repository: https://huggingface.co/camenduru/unianimate/tree/main and download the five checkpoints you find there.

Also visit this huggingface repository to download the animate-x checkpoint: https://huggingface.co/Shuaishuai0219/Animate-X/tree/main

You only need to download the animate-x_ckpt.pth checkpoint. All other checkpoints are the same as those downloaded from the huggingface repository first mentioned.

Place the checkpoints in your ' ComfyUI \custom_nodes\ComfyUI-UniAnimate-W-main\checkpoints' folder.

Close your ComfyUI terminal and launch it again. You should now be able to use the nodes.

 

Comments

Post a Comment

Popular posts from this blog

Running Animate-X on Google Colab for Free

Image
Introduction Animate-X is an image animation AI tool for animating an image using a driving video. Many believe that running advanced AI animation tools like Animate-X requires high-end hardware. However, with the right optimizations, you can harness the power of Google Colab’s free T4 GPU to bring animations to life. This guide walks you through every step, from accessing the Animate-X Colab notebook to configuring the best settings for smooth performance. Whether you're a creator experimenting with AI driven animations or simply exploring new tools, this tutorial will help you get started efficiently. If you prefer a video tutorial, then you can watch the video below. Accessing the Animate-X Colab Notebook The first step is to visit the Google Colab Notebooks GitHub repository  It will appear as shown below. Next, locate the Animate-X Colab notebook and click on 'Open in Colab' to launch it as shown below.  This will open the notebook in Google Colab, where all the nece...
Read more

Derivation of the Loan Repayment Formula

I once took a loan of #1,000,000 at an interest rate of 24% per annum (or 2% monthly) from my bank. I assumed at the time that I would end up paying the bank a total of #1,240,000, with the extra #240,000 being the total interest. I expected to pay the sum of #103,333.33 monthly, with my last payment being at the end of the 12 th month. My calculation was as shown below; Assuming Loan taken = L, total interest = ti, annual interest rate = i,  Total amount to be paid = T, and monthly payments = m, T = L + ti T = L + L x i  (where ti = L x i) Recall that L = 1000000 and i = 24% So T = 1000000 + 1000000 x 0.24 = #1,240,000 Therefore, my monthly repayments will be, m = #1,240,000/12 = #103,333.33. However, at the end of the first month, my bank deducted approximately #94,559.60 from my account. At first, I thought they made a mistake. But then I imagined that these money-hungry entities might make mistakes crediting an account below the expected amount, but it is almost impossibl...
Read more