Flux Krea Dev: Photorealistic Portraits Without the AI Look - Workflow + Guide
9 min read
Flux Krea Dev: Photorealistic Portraits Without the AI Look - Workflow + Guide
Prompt: A realistic person wearing a bright yellow raincoat stands smiling near the right edge of a wide, landscape-oriented frame. The background is a vibrant urban skatepark, with colorful ramps and graffiti, set after a rain—puddles reflecting bold street art and cloudy sky. The person has authentic skin texture and natural facial features, fully clothed (no topless or naked appearance), and holds a skateboard with one hand. The words "Flux Krea" are clearly visible as part on the clothing. Lighting is soft and natural, with realistic shadows and reflections. No plasticky skin, no overly smooth surfaces, no AI artifacts, no uncanny valley. The composition is dynamic and wide, blending urban energy with lifelike detail. Highly detailed, organic, and human appearance
Flux Krea Dev is built to create photorealistic images without the usual AI giveaways-no plastic skin, no overly smooth textures, no uncanny valley weirdness.
This is a collaboration between Black Forest Labs and Krea AI. It's a 12-billion parameter model that handles realistic skin tones, natural lighting, and follows prompts accurately. Works especially well for portraits and character shots.
Faster than standard Flux Dev and produces more varied, lifelike results.
Flux Krea Dev is designed to generate photorealistic images while avoiding common AI artifacts like plastic-looking skin or overprocessed textures. The 12-billion parameter model maintains realistic detail, handles nuanced lighting well, and produces natural skin tones.
It supports fine-tuning, works with existing Flux workflows, and generates images quickly. Good for portraits, character work, and any project where you need genuinely realistic human features.
Comparison of Flux Krea Dev with Standard model
Prompt: A female assassin dances fluidly atop a city rooftop at night, her sleek, dark attire blending modern tactical gear with elegant, flowing elements. Neon lights from the city skyline reflect off her outfit as she moves with precision and grace, her silhouette striking against the urban backdrop. Seed - 521404981828559, Euler, Simple, Steps 25, CFG 1
Flux Krea Dev is a major upgrade from the original Flux Dev, designed to create genuinely lifelike images. Where Flux Dev often produces flat, repetitive, or generic results, Flux Krea Dev excels in capturing photorealistic detail, prompt accuracy, and expressive variety—making it the best open-source choice for anyone seeking high-quality, realistic AI-generated art and portraits.
Get ready for a hands-on experience designed to make your art leap off the screen and make people ask, “Is that really AI?”
Verified to work on ThinkDiffusion Build: June 27, 2025
ComfyUI v0.3.47 with the use flux1-krea-dev.safetensors
Note: We specify the build date because ComfyUI and custom node versions updated after this date may change the behavior or outputs of the workflow.
Minimum Machine Size: Ultra
Use the specified machine size or higher to ensure it meets the VRAM and performance requirements of the workflow
💡
Download the workflow and drag & drop it into your ComfyUI window, whether locally or on ThinkDiffusion. If you're using ThinkDiffusion, minimum requirement is the Turbo 24gb machine, but we do recommend the Ultra 48gb machine.
Custom Nodes
If there are red nodes in the workflow, it means that the workflow lacks the certain required nodes. Install the custom nodes in order for the workflow to work.
Go to the ComfyUI Manager > Click Install Missing Custom Nodes
Check the list below if there's a list of custom nodes that needs to be installed and click the install.
Required Models
For this guide you'll need to download these 4 recommended models.
Search for the models above and when you find the exact model that you're looking for, click install, and make sure to press refresh when you are finished.
If Model Manager doesn't have them: Use direct download links (included with workflow) and upload through ThinkDiffusion MyFiles > Upload URL. Refer our docs for more guidance on this.
You could also use the model path source instead: by pasting the model's link address into ThinkDiffusion MyFiles using upload URL.
Model Name
Model Link Address
ThinkDiffusion Upload Directory
flux1-krea-dev.safetensors
📋 Copy Path
.../comfyui/models/diffusion_models/
t5xxl_fp32.safetensors
📋 Copy Path
.../comfyui/models/clip/
clip_l.safetensors
📋 Copy Path
.../comfyui/models/clip/
ae.safetensors
📋 Copy Path
.../comfyui/models/vae/
💡
If the flux1-krea-dev.safetensors is already a pre-loaded model, you don't need to upload the model anymore.
Step-by-step Workflow Guide
This workflow was pretty easy to set up and runs well from the default settings. Here are a few steps where you might want to take extra note.
Steps
Recommended Nodes
1. Set the Models
Set the models as seen on the image.
2. Write a Prompt
Write a prompt that describes a realistic description. It is better to depict a character or portrait inthe prompt.
3. Check the Sampling
Set the settings as seen on the image. Don't change the scheduler or sampler as it may result to bad quality.
4. Check the Generated Image
Examples
Prompt: A vivid portrait of a modern street artist, spray paint stains on their hands and a colorful bandana covering part of their face. The character stands against a bright graffiti wall, with energetic splashes of neon paint in the background. Their eyes are lively and creative, capturing urban spirit and bold individuality. Studio-quality lighting, crisp detail, portrait formatPrompt: A close-up portrait of a mysterious elf scholar, pointed ears visible beneath tousled silver hair, intricate crystal earrings, and a deep blue cloak embroidered with ancient runes. The character is posed against a backdrop of weathered library shelves filled with glowing magical tomes. Soft, moody lighting highlights delicate facial features and thoughtful eyes. Highly detailed, fantasy avatar style, portrait orientationPrompt: A close-up shot of a young woman with curly light brown hair. The woman's hair is blowing in the wind and cascades over her shoulders. She is wearing a white tank top with a white strap on the back. Her eyes are open and her lips are pursed. The background is blurred out, but it appears to be a sunny day. The sky is a light blue, and the sun is shining down on the left side of the image.Prompt: A realistic person shown from the waist up in a landscape-oriented image, positioned randomly within the frame (off-center, left, right, or anywhere unexpected). The person is fully clothed in casual, modern attire—no topless or naked appearance. The background is visually interesting and randomly chosen, such as a lively city street, tranquil beach, cozy café, lush garden, misty park, or an abstract blur. The person has authentic skin texture, natural facial features, and expressive emotion, with realistic lighting and true color tones. No plasticky skin, no overly smooth surfaces, no AI artifacts, no uncanny valley. Composition is wide and balanced, with distinct background elements contributing to the atmosphere. Highly detailed, organic, human appearancePrompt: An indoor shot of a man smoking a cigarette. The smoke is coming from behind his head, obscuring his face in the upper right corner of the frame. His left hand is resting on his chin, while his right hand is holding the cigarette in his left hand. His mouth is slightly open, as if he is about to smoke the cigarette. His right ear is visible, and his upper left ear is showing. His eyes are dark, and he has a small amount of light shining on his face. The lighting is subdued, as evidenced by the shadow of the man's face on the right side of the image.Prompt: A medium-sized woman stands in front of a colorful graffiti wall. The woman's hair is long and cascades down to her shoulders. She is wearing a black short-sleeved t-shirt and black pants. The word "FLOYO" is written in large, bold letters in a vibrant shade of pink, orange, and black. The letters are outlined in a darker shade of black. The wall behind the woman is a vibrant combination of blue, green, yellow, and orange. The ground beneath her is a dark gray asphalt.Prompt: A highly realistic portrait of a person with authentic skin texture and natural facial features, expressive eyes, and genuine emotion. The background is random and visually interesting—such as an urban street, cozy café, lush garden, or abstract blurred scene—adding character to the composition. Lighting is soft and flattering, with true-to-life colors and gentle shadows. No plasticky skin, no AI artifacts, no overly smooth surfaces, no uncanny valley. Rich detail, organic appearance
Troubleshooting
Red Nodes: Install missing custom nodes through ComfyUI Manager Out of Memory: Use smaller expansion factors or switch to Ultra machine Poor Quality: Check input image resolution and adjust kontext strength Visible Seams: Lower strength and ensure good prompt description
If you’re having issues with installation or slow hardware, you can try any of these workflows on a more powerful GPU in your browser with ThinkDiffusion.
Member discussion