# 🔥 ONE MEMBERSHIP = FULL ACCESS
Your membership includes :
All models
Exclusive LoRAs
Exclusive workflows
Scripts & tools
All future updates included
⭐ Get every current model, every future release, and all exclusive content with a single membership.
📢 Join The Community
We're building a friendly space to chat, share creations, and get support.
👉 Click here to join the Discord
Krea 2 Ultimate Workflow V1
Turn Krea 2 into a complete production-ready image generation system.
This workflow combines a powerful Prompt Builder, LoRA support, optional NSFW optimization, and an integrated SeedVR2 upscaling pipeline to maximize image quality while minimizing prompt engineering.
🚀 Features
🧠 Advanced Prompt Builder
Stop writing long prompts manually.
The integrated Prompt Builder lets you create highly detailed prompts through a simple interface.
Features:
Subject description
Background description
Photography presets
Lighting controls
Aesthetic controls
Resolution presets
JSON import/export
Ideogram-style editor
Perfect for both beginners and advanced users who want faster, more consistent results.
🎨 LoRA Support
Easily load and combine your favorite LoRAs directly inside the workflow.
Perfect for:
Characters
Styles
Clothing
Concepts
Custom trained models
Designed to work seamlessly with Krea 2.
🔥 Optional NSFW Enhancement
Built-in conditioning controls designed to improve generation quality when working with NSFW content.
Benefits:
Better anatomy
More consistent body proportions
Improved skin rendering
Cleaner generations
The multiplier is set to 0.5 by default. The more you increase it, the more NSFW content will appear. But you risk losing some realism.
⚡ SeedVR2 Upscaling Pipeline
Integrated SeedVR2 enhancement system.
Features:
Face enhancement
Detail recovery
Sharper eyes and hair
Better texture reconstruction
Improved clothing details
Higher perceived resolution
Generate first, upscale automatically, and get a polished final image without leaving the workflow.
🎛 One-Click Options
Quick toggles for:
Prompt Builder
Image Preview
NSFW Conditioning
SeedVR2 Upscaling
Enable only what you need.
Description
Krea2 WorkFlow V2.0
There are many changes in this Version II.
After several tests, I haven't been able to find a better method than this one so far.
A simple and effective upscaler has been added, which is much faster for systems with limited VRAM.
Two samplers for a higher-quality initial output.
FaceDetailer has been adjusted to fine-tune faces slightly, particularly the eyes and mouth.
The choice of upscaler model has a significant impact on the photo. It's up to you to choose the one you like best.
This WF is ready for Krea IntoRealism, which will be released in the coming days.
The images were generated using the base model, without any LoRA.
FAQ
Comments (7)
This WF works great, thanks for sharing it.
Without Loras this generates the image within 50 seconds. If I use a lora, it will take forever. Seems to stuck in First Pass. Tried several times with different loras but this seems not to work with loras.
No one has reported this issue to me. Let's see if everything is OK with your Comfyui installation.
@enzino Could it be that my 16GB GeForce 4080 is the limiting factor? The Checkpoint has 12.5 GB. So maybe I run out of VRam if I use an additional lora?
@enzino That’s sorted now, and it was entirely my fault. My CUDA wasn’t up to date. I’d somehow missed a downgrade. After updating PyTorch to a version with CUDA 13.0, everything’s working fine.
Sorry, it was my fault!