Simple Comfyui Workflow with controlnets
Simple workflow to convert a toon to real. I did not create this. Controlnet is Union XL
Runtime profile
- Access
- External workflow
- Status
- Listed
- Source
- Civitai
- Updated
- Oct 25, 2026
Source description
Simple workflow to convert a toon to real. I did not create this. Controlnet is Union XL
V1. : Simple flow, image to image using the control net. Be sure to put in prompts to help get it closer to what you want to see.
V2: I added in a face detailer and a lora. This gives you a little boost especially if you're looking to apply a character lora to the output.
What you need
Hardware and download-size estimates are not available yet.
Requirements
Models
6- Not resolved
Object Detector · Hugging Face · bbox/face_yolov8m.pt
centerfold_v90Naughty.safetensors
Not resolvedCheckpoint · Unknown
controlnet++_union_sdxl.safetensors
PossibleControlNet · Unknown
detailer_v5-SDXL.safetensors
Not resolvedLoRA · Unknown
epicrealismXL_vxFinalkiss_969698.safetensors
PossibleCheckpoint · Unknown
VAEDecode
PossibleVAE · Unknown
Node packs
14Node pack · Unknown
- Found
Node pack · Unknown
CheckpointLoaderSimple
PossibleNode pack · Unknown
CLIPTextEncode
PossibleNode pack · Unknown
ControlNetApply
PossibleNode pack · Unknown
ControlNetLoader
PossibleNode pack · Unknown
Node pack · Unknown
EmptyLatentImage
PossibleNode pack · Unknown
- Found
Node pack · Unknown
KSampler
PossibleNode pack · Unknown
LoadImage
PossibleNode pack · Unknown
PreviewImage
PossibleNode pack · Unknown
SaveImage
PossibleNode pack · Unknown
UltralyticsDetectorProvider
PossibleNode pack · Unknown
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Purpose
This workflow converts a toon image toward a real-style image with image-to-image processing and Union XL ControlNet, using an image and prompts to produce an image output.
Flow
V1 is the simple image-to-image version using ControlNet.
V2 adds a face detailer and a LoRA, with the description highlighting character-LoRA use on the output.
Inputs and outputs
The workflow names LoadImage, PreviewImage, and SaveImage nodes for loading, previewing, and saving the image.
Setup
It is listed under Workflows, with SDXL 1.0 as the base model and Scorpiov2 as publisher.
Dependencies
Custom node packs listed for this workflow are ComfyUI-Impact-Pack (FaceDetailer), ComfyUI-Impact-Subpack (UltralyticsDetectorProvider), comfyui-art-venture (AV_LoraLoader and AV_ControlNetPreprocessor), and ComfyUI_Comfyroll_CustomNodes (CR Image Input Switch (4 way)).
Other named nodes are CheckpointLoaderSimple, CLIPTextEncode, ControlNetApply, ControlNetLoader, EmptyLatentImage, KSampler, and VAEDecode.
The listed model files are bbox/face_yolov8m.pt, centerfold_v90Naughty.safetensors, controlnet++_union_sdxl.safetensors, detailer_v5-SDXL.safetensors, and epicrealismXL_vxFinalkiss_969698.safetensors.
Configuration
Add prompts to steer the result toward what you want to see.
Suggestion · not verified
Consider V2 if you want to apply a character LoRA to the output; it adds a face detailer and a LoRA.
Suggestion · not verified
Cautions
Before use, verify that the five listed model files are available in your setup.
Suggestion · not verified
The listing does not specify a license.
Does this need to be edited?
Sign in to send an edit request.
Sources
1 source
Sources
1 sourceSource excerpts
1 excerpt
Source excerpts
1 excerptSource context: 263 downloads · Type Workflows · Base model SDXL 1.0
