Skin Perfection Unleashed: A Comprehensive FLUX Workflow Tutorial
Unlock FLUX-based text-to-image workflow for stunning skin texture enhancement & detail refinement. Discover the power of efficient generation, face/body inpainting, and cinematic color grading. Dive in and elevate your image creation!
- Use Case
- Text In Image
- Best For
- Text In Image
- Models
- FluxControlnetLora
- Key Nodes
- ControlnetUpscalerInpaint
- VRAM
- Medium VRAM (12–16GB)
- Reading Time
- 3 min
Workflow Overview
Unlock FLUX-based text-to-image workflow for stunning skin texture enhancement & detail refinement. Discover the power of efficient generation, face/body inpainting, and cinematic color grading. Dive in and elevate your image creation!
Content type: Workflow
Primary intent: Download
Required Models
- Flux
- Controlnet
- Lora
Required Nodes
- Controlnet
- Upscaler
- Inpaint
Setup Notes
- Install the required models before opening the workflow template.
- Recommended hardware: Medium VRAM (12–16GB).
1. Workflow Overview
This is a FLUX-based text-to-image workflow specialized in skin texture enhancement and detail refinement. Key features:
Uses FP8-quantized FLUX model for efficient generation
Integrates Ultralytics detector + SAM for face/body inpainting
Applies LUT color grading and super-resolution
Core Models:
FLUX (flux1-dev-fp8.safetensors): Base FP8-quantized model
SD3 CLIP (clip_l + t5xxl_fp8): Dual-text encoder for prompt understanding
LoRA (FLUX太强悍lora): Skin detail fine-tuner
ControlNet (Eyeful_v2-Paired.pt): Facial feature control
2. Critical Nodes
Node | Function | Installation |
|---|---|---|
| Custom diffusion sampling | Requires |
| Face/body detection | Via |
| Super-resolution | Install |
| Cinematic color grading | Requires |
Dependencies:
Model Files:
Place
flux_ae.sftinmodels/vaeDownload
x1_ITF_SkinDiffDetail_Lite_v1.pthtomodels/upscale_models
Plugins: Must install
Impact PackandComfyUI-Manager
3. Workflow Structure
Group Logic:
Text Encoding (Top-Left):
Input: Fantasy-style prompt + negative prompt
Output: CLIP conditioning
FLUX Generation (Center):
Input: 1024x1024 latent + Euler sampler
Output: Raw image
Detail Refinement (Bottom-Right):
Pipeline: Face detect → Inpaint → 2x upscale → LUT grading
Key params: Denoise 0.3, Steps 20
4. Inputs & Outputs
Required Inputs:
Positive prompt: Must include skin descriptors (e.g., "porcelain skin")
Resolution: Fixed at 1024x1024 (FLUX constraint)
Seed: Randomizable
Final Output:
Format: PNG with metadata
Processing chain: Raw → Sharpen → 2x upscale → LUT
5. Notes
VRAM: ≥12GB required (even with FP8)
Common Errors:
LUT not found: VerifyPB_Boulder.CUBEpathSAM load fail: Downloadsam_vit_b_01ec64.pth
Optimization:
Enable
TAESDdecoder for faster previewReduce
CR Upscaletile size to 512 for lower VRAM