β¨ Illustrious Ultimate Upscaler | Img2Img + LoRA Support
Take your Illustrious generations to the next level with this ComfyUI workflow designed for high-quality Img2Img upscaling and refinement while maintaining full LoRA compatibility.
Whether you're working with portraits, characters, illustrations, or highly detailed anime artwork, this workflow helps recover fine details, improve overall image quality, and enhance visual consistency without sacrificing the original composition.
Features
β
Img2Img-based upscaling and refinement
β
Full LoRA support
β
Preserves character identity and style
β
Improves details, textures, and sharpness
β
Great for anime, illustrations, and character art
β
Flexible settings for different upscale strengths
β
Easy to customize and expand
Recommended Use Cases
Enhancing low-resolution generations
Refining existing Illustrious outputs
Adding extra detail before final upscaling
Improving facial features and clothing details
Preparing images for sharing, printing, or further editing
Tips
Lower denoise values help preserve the original image.
Higher denoise values allow for stronger refinement and detail generation.
Experiment with different LoRAs to achieve unique artistic styles.
For best results, start with a reasonably clean source image.
Compatible With
Illustrious-based checkpoints
Character LoRAs
Style LoRAs
Detail LoRAs
Most standard ComfyUI installations
Notes
This workflow was built to provide a balance between detail enhancement, image stability, and LoRA flexibility. Results may vary depending on the checkpoint, LoRAs, prompts, and source image quality.
If you find it useful, consider leaving a rating and sharing your results. Seeing what others create with the workflow is always appreciated. π
Version: Initial Release
Workflow Type: ComfyUI Img2Img Upscaler
Model Family: Illustrious
LoRA Support: Yes β
Description
This version is for processing all images in the directory.
FAQ
Comments (3)
I'm having a problem with this program. When I click run, after setting everything up, I get an error message saying "TypeError: 'NoneType' object is not subscriptable", and it has something to do with the preview image.
Does anyone know what's going on?
Thank you in advance!
Check if any nodes are outlined in red and ensure all custom nodes are installed and working; errors can occur if the file path for the images to be upscaled is incorrect. Copy the exact directory path where the images are located from your file explorer. I can provide a quick video showing how the workflow works if needed.
@GilsonAnimesΒ The video would be greatly appreciated, thank you! I downloaded the checkpoints, LoRAs, and upscale models, but I don't know what else I have to do.
I'm new to ComfyUI and Civitai, forgive me.


