Download
1 variant available
This checkpoint includes a config file, download and place it along side the checkpoint.
Minor Improvements

2.7K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
9.4K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
License:
AnimaThe Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
Built on NVIDIA Cosmos
V4 Update :
Added Metadata Saver for easy CivitAi uploads : You won't have to add resources manually anymore
Added Group bypasser : Now You can enable/disable controlnet or facedetailer utilites by one click
Downloading Metadata Saver :
Open your ComfyUI/custom_nodes folder, type CMD in the address bar and run:
git clone https://github.com/xxmjskxx/ComfyUI_SaveImageWithMetaDataUniversalAfter installation, restart ComfyUI and load the workflow.
V3 Update :
Added a FaceDetailer node. It detects, re-renders, and re-attaches faces to the generated image, resulting in sharper and more detailed faces overall.
You can Increase
bbox_thresholdanddrop_sizeto reduce detection of small/background faces.All detectors and the SAM model come with the node once you install it from the ComfyUI Library, so you won't need any external resources.
V2 Update :
Since the Anima ControlNet node was recently updated, the older version will no longer work if you update your ComfyUI.
So, I had to update this workflow to make it compatible with the latest version.
ControlNet has added a new "Model Patch" input. I've added the required node, but you will need to download the necessary models and place them in your:
ComfyUI/models/model_patches/Download them from here:
https://huggingface.co/Comfy-Org/Anima-LLLite/tree/main/model_patches
Additional Improvements
Added a Size Selector to easily switch between popular image sizes and flip the dimensions.
Added an Image Comparer (On top the Save Image node)
Added a Stack LoRA Loader.
Moved a few things around to make the workflow more compact and easier on the eyes.
If You don't need the controlnet just select the yellow nodes and press Ctrl+B to bypass them , then it would be a simple Text to Image workflow.
A small, clean, and lazy-friendly Anima workflow for those who don't want to play "Where is that damn node?" every time they want to change something.
I made this workflow to keep the important stuff close together:
✅ Easy checkpoint switching
✅ Quick character/model changes
✅ Simple prompt editing
✅ Fast sampler and generation setting adjustments
✅ Less scrolling, less node hunting, more creating
The idea is simple: everything you actually touch while experimenting is right in front of you. No giant spaghetti workflow with 500 nodes
Just open it, pick your model, change your character/prompt, tweak your settings, and generate.
Required Model:
You need to download the Anima ControlNet-LLLite model first and place it this path:
ComfyUI/models/controlnet
Download:
https://huggingface.co/kohya-ss/Anima-LLLite/tree/main
This workflow uses the Anima LLLite ControlNet setup, so make sure the model is installed before running it.
Quick heads-up: CivitAI did that funny thing before where my other workflow straight-up disappeared into the void. 💀
I’ve uploaded it again, but if this happened to this one once more , just grab one of the images, they contain the workflow.
Have fun creating!


