Workflow - 3 different characters in one picture — v1.0
With this workflow you can modify 3 different character with your reference image character you prefer. PULID Flux can detech face, you can config in option node.
Runtime profile
- Access
- External workflow
- Status
- Listed
- Version
- [Flux]_[PULID]_3-face.json
- Source
- Civitai
- Updated
- Oct 18, 2026
Source description
With this workflow you can modify 3 different character with your reference image character you prefer. PULID Flux can detech face, you can config in option node.
Note: Detail Daemon it optional. disabled this group it will help the processing to be faster.
Work based on.
resolution width 1024px ( Another size you must calculate Mark by yourself )
What you need
- Model files
- 8.9 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Requirements
Models
6ae.safetensors
PossibleVAE · Unknown
clip_l.safetensors
PossibleText encoder · Unknown
- Found
Upscaler · 148 MB · Hugging Face · nguoidoncui/ComfyUI
Checkpoint · 7.72 GB · GGUF · Unknown
Checkpoint · 1.06 GB · SAFETENSORS · Unknown
t5-v1_1-xxl-encoder-Q5_K_S.gguf
Not resolvedText encoder · Unknown
Node packs
11cg-use-everywhere
FoundNode pack · Registry
comfy-image-saver
FoundNode pack · Registry
comfyui_essentials_mb
FoundNode pack · Registry
comfyui_patches_ll
FoundNode pack · Registry
comfyui_pulid_flux_ll
FoundNode pack · Registry
comfyui_ultimatesdupscale
FoundNode pack · Registry
comfyui-detail-daemon
FoundNode pack · Registry
comfyui-easy-use
FoundNode pack · Registry
ComfyUI-GGUF
PossibleNode pack · Registry
- Found
Node pack · Registry
rgthree-comfy
FoundNode pack · Registry
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Purpose
This workflow is for modifying an image with three different characters: it uses a loaded image and a preferred character reference image, applies PuLID Flux face detection with configurable options, and outputs a modified picture.
Flow
The 66-node workflow includes face detection, image concatenation, previews, comparisons, saving, image scaling, and custom upscaling.
Inputs and outputs
The workflow includes a LoadImage stage and uses a preferred character reference image as inputs.
It includes PreviewImage, Image Comparer (rgthree), and comfy-image-saver stages for previewing, comparing, and saving image results.
Setup
Open the v1.0 workflow file named [Flux]_[PULID]_3-face.json.
Suggestion · not verified
Dependencies
The listed node packs are cg-use-everywhere, comfy-image-saver, comfyui_essentials_mb, comfyui_patches_ll, comfyui_pulid_flux_ll, comfyui_ultimatesdupscale, and comfyui-detail-daemon.
The remaining listed node packs are comfyui-easy-use, ComfyUI-GGUF, comfyui-kjnodes, and rgthree-comfy.
The listed model files are ae.safetensors, clip_l.safetensors, DAT_x4.pth, flux1-dev-Q5_K_S.gguf, pulid_flux_v0.9.1.safetensors, and t5-v1_1-xxl-encoder-Q5_K_S.gguf.
Configuration
Use PulidFluxOptions to configure face-related options.
Suggestion · not verified
Detail Daemon is optional, so its group can be disabled.
Suggestion · not verified
Use 1024 px as the stated working width; for another size, calculate the setting separately.
Suggestion · not verified
Cautions
Before setup, verify the exact file or package for ae.safetensors, DAT_x4.pth, pulid_flux_v0.9.1.safetensors, t5-v1_1-xxl-encoder-Q5_K_S.gguf, and ComfyUI-GGUF.
Suggestion · not verified
The workflow package does not include the six model files, so plan to provide them separately.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 source
Sources
1 sourceSource excerpts
2 excerpts
Source excerpts
2 excerptsSource context: 224 downloads · Type Workflows · Base model Flux.1 D
Source context: 189 downloads · Type Workflows · Base model Flux.1 D
