Dirty Realism Krea2 UNCENSORED!!!!
About this version
Dirty Realism Krea2 V3
in this version 3, I uploaded the INT8 and INT8-CONVROT quantized models, as well as the FP8 version.
Version 3 is focused on one goal: bringing AI-generated images even closer to real photography.
This update delivers a major improvement in high-resolution generations, producing cleaner results with significantly fewer artifacts while preserving the raw, documentary-style aesthetic that defines Dirty Realism. The grain has been carefully refined, making it feel more natural and photographic instead of distracting, resulting in images that are cleaner without losing their authentic character.
Overall realism has taken another step forward. Materials, lighting, skin, textures, and environmental details now appear more convincing, creating images that feel less like AI generations and more like genuine photographs captured in the real world.
What's Improved
Significantly better performance at higher resolutions.
Major reduction of visual artifacts and unwanted distortions.
Refined photographic grain for a cleaner, more natural look.
Improved realism across lighting, materials, and textures.
More believable photographic results while preserving the signature raw aesthetic.
Better overall consistency across a wide variety of prompts.
This release comes shortly after Version 2 for a reason. Over the past days, the model has been refined through up to 10 hours of daily testing, training, and fine-tuning, allowing rapid improvements to be validated and released as soon as they proved consistently better. Rather than waiting months between updates, the goal is to deliver meaningful improvements to the community as soon as they're ready.
Dirty Realism Krea2 V3 is the most polished and realistic version yet, delivering cleaner images, stronger high-resolution performance, and an even more convincing photographic look while preserving the imperfect authenticity that makes real photography feel alive.
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 5 denois 0.10 Hires CFG 1,
Width and height: 896*1152 or Hires 1152*1440
CFG Scale: 1
Steps: 8
The sampling method i use is ER SDE-Simple, but you can try a few more, such as Euler-Beta Uniform or Euler A-Simple
Important update: in this version 2, I uploaded the INT8 and INT8-CONVROT quantized models, as well as the FP8 version.
About V2
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 5 denois 0.20 Hires CFG 1,
Width and height: 896*1152
CFG Scale: 1
Steps: 8
The sampling method i use is ER SDE-Simple, but you can try a few more, such as Euler-Beta Uniform or Euler A-Simple
Dirty Realism Krea2 V2
The next evolution of authentic AI photography is here.
Dirty Realism Krea2 V2 is not just an update. It's a complete refinement of the philosophy behind the model. Every aspect has been revisited with one goal in mind: making AI images feel less artificial and more like genuine photographs captured in the real world.
Version 2 pushes realism further than ever before. The model produces richer textures, more convincing lighting, natural skin, authentic facial features, and a level of environmental detail that makes every image feel lived in rather than rendered. Instead of chasing flawless beauty, it captures the subtle imperfections that our eyes instinctively associate with reality.
The signature raw aesthetic has been significantly enhanced. Images now feature more believable camera behavior, realistic exposure, organic shadows, subtle sensor noise, natural motion blur, imperfect framing, and the tiny inconsistencies that transform an AI generation into something that could easily be mistaken for a real photograph.
What's New in Version 2
Dramatically improved raw photographic realism.
More natural skin tones with realistic pores, texture, and imperfections.
Better facial anatomy and human expressions.
Enhanced lighting with richer highlights, softer transitions, and realistic shadows.
Superior material rendering including fabric, concrete, metal, glass, and weathered surfaces.
Increased environmental storytelling with scenes that feel naturally inhabited.
Improved depth, atmosphere, and cinematic composition.
More convincing handheld camera aesthetics and documentary-style photography.
Better consistency across a wide variety of prompts.
Stronger performance in portraits, street photography, urban exploration, interiors, lifestyle, fashion, industrial scenes, and cinematic storytelling.
Why Download Dirty Realism Krea2 V2?
If you've ever felt that AI images look too clean, too polished, or instantly recognizable as synthetic, this model was created specifically to solve that problem.
Dirty Realism Krea2 V2 focuses on authenticity over perfection. It embraces the subtle chaos of real photography: imperfect lighting, spontaneous moments, natural textures, environmental wear, and the visual imperfections that make images emotionally believable.
Whether you're creating cinematic scenes, documentary photography, fashion editorials, architectural visualization, urban environments, or everyday moments, this model consistently delivers images with remarkable realism and character.
Every generation feels like it has a story behind it.
Not because it tries to imitate reality.
Because it understands what makes reality convincing.
Download Dirty Realism Krea2 V2 and experience a model designed for creators who believe the most realistic images aren't the perfect ones... they're the ones that feel impossible to fake.
Most AI image models chase perfection. Dirty Realism Krea2 does the opposite. It embraces the beautiful imperfections that make photographs feel alive. Tiny flaws, natural lighting, realistic textures, imperfect framing, authentic skin, motion blur, sensor noise, and candid compositions all come together to create images that look captured, not generated.
Built on Krea2, this model is designed to produce scenes with genuine atmosphere instead of sterile perfection. Every image feels like it has a story behind it, whether it's a spontaneous street photo, an old smartphone snapshot, a documentary frame, or a cinematic moment frozen in time.
Why you'll love it:
Natural, believable realism instead of overprocessed AI aesthetics.
Exceptional lighting and texture reproduction.
Authentic skin, clothing, and environmental details.
Beautiful imperfections that increase photographic credibility.
Strong performance for portraits, street photography, documentary, lifestyle, urban scenes, and cinematic compositions.
Consistent results while maintaining a raw, organic look.
If you're tired of images that scream "AI-generated," Dirty Realism Krea2 was built for you. It delivers photographs with personality, atmosphere, and the subtle imperfections that convince the eye.
Download it, experiment with it, and discover how realistic AI imagery can become when perfection is no longer the goal.
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 5 denois 0.20 Hires CFG 1,
Width and height: 896*1152
CFG Scale: 1
Steps: 8
The sampling method i use is Euler-Beta, but you can try a few more, such as ER SDE-SGM Uniform or Euler A-Simple
Since it is a turbo model, it is a good idea to use the recommended parameters.
I’ve never really liked putting my models behind early access, but if you’re able to donate some Buzz, it’s appreciated; if not, feel free to enjoy the model anyway. I really look forward to seeing your creations. This is my first Krea2 model, so I imagine there might be a few flaws that we hope to iron out over time.
Description
Dirty Realism Krea2 V2
The next evolution of authentic AI photography is here.
Dirty Realism Krea2 V2 is not just an update. It's a complete refinement of the philosophy behind the model. Every aspect has been revisited with one goal in mind: making AI images feel less artificial and more like genuine photographs captured in the real world.
Version 2 pushes realism further than ever before. The model produces richer textures, more convincing lighting, natural skin, authentic facial features, and a level of environmental detail that makes every image feel lived in rather than rendered. Instead of chasing flawless beauty, it captures the subtle imperfections that our eyes instinctively associate with reality.
The signature raw aesthetic has been significantly enhanced. Images now feature more believable camera behavior, realistic exposure, organic shadows, subtle sensor noise, natural motion blur, imperfect framing, and the tiny inconsistencies that transform an AI generation into something that could easily be mistaken for a real photograph.
What's New in Version 2
Dramatically improved raw photographic realism.
More natural skin tones with realistic pores, texture, and imperfections.
Better facial anatomy and human expressions.
Enhanced lighting with richer highlights, softer transitions, and realistic shadows.
Superior material rendering including fabric, concrete, metal, glass, and weathered surfaces.
Increased environmental storytelling with scenes that feel naturally inhabited.
Improved depth, atmosphere, and cinematic composition.
More convincing handheld camera aesthetics and documentary-style photography.
Better consistency across a wide variety of prompts.
Stronger performance in portraits, street photography, urban exploration, interiors, lifestyle, fashion, industrial scenes, and cinematic storytelling.
Why Download Dirty Realism Krea2 V2?
If you've ever felt that AI images look too clean, too polished, or instantly recognizable as synthetic, this model was created specifically to solve that problem.
Dirty Realism Krea2 V2 focuses on authenticity over perfection. It embraces the subtle chaos of real photography: imperfect lighting, spontaneous moments, natural textures, environmental wear, and the visual imperfections that make images emotionally believable.
Whether you're creating cinematic scenes, documentary photography, fashion editorials, architectural visualization, urban environments, or everyday moments, this model consistently delivers images with remarkable realism and character.
Every generation feels like it has a story behind it.
Not because it tries to imitate reality.
Because it understands what makes reality convincing.
Download Dirty Realism Krea2 V2 and experience a model designed for creators who believe the most realistic images aren't the perfect ones... they're the ones that feel impossible to fake.
FAQ
Comments (31)
I dont know what is wrong but generated images are washed out.
And with half of loras it becomes also very overlit and messed like drawing made of color spots.
Is model incompatible with loras?
Any chance for an INT8 ConvRot version for people that are using a 30X0 GPU?
OwO
I'm working on that; I hope to publish it soon.
All set, my friend—the INT8 version and the INT8-CONVROT version are up.
@Sateluco But can we please get bf16 too? :(
@Sateluco UwU thank you! >w<
Another bug i see - model very biased to close up on face.
Literally impossible to generate normal full height view, its always from above close up on face. When i try to describe full height details it just becomes fish eye.
To be honest i not yet understand what great other people see here, model not working properly and not follow prompts.
I tested it with prompts used for another models so prompts are definetely fine - but this model not follow them.
@velanteg I don't have your problems, I wonder why??
Fascinating theory, but everyone else seems to be getting full-body shots just fine. Have you considered that maybe the model isn't broken, but your prompt engineering needs an upgrade?
@Tukanazo1966 my prompt works with all another models. So its your model without doubt.
@Tukanazo1966 I posted example what generated by normal models and what creating yours with same prompt.
On site posting is broken so you can see it in NSFW showcase in civitai discord.
@velanteg Its My model My friend, i listen what do You have to Say in order to improve
@velanteg I have a question. Did you describe the entire subject in detail, or mainly the face?
Krea seems to respond much better to natural, descriptive prompts than to prompts originally written for other models. Instead of only writing "full body woman", I described the face, body proportions, posture, clothing, camera perspective and environment in full sentences.
I'm not saying this is the solution in every case, but in my testing the checkpoint works very well. The example below generated a proper full-body composition without any fisheye distortion.
Maybe it's worth trying a prompt written specifically for Krea instead of reusing one designed for a different model.
***************************
a simple example
Canon EOS 6D
Lens: Canon EF 50mm f/1.8
Focal length: 50mm
Aperture: f/2.0
Shutter speed: 1/250s
ISO: 200
White balance: daylight
Lighting: natural window light
Depth of field: shallow DOF
Look: realistic skin tones, soft shadows, natural full-body lifestyle photography.
A realistic full-body photograph of a beautiful 24-year-old Taiwanese woman. Her complete figure is clearly visible from head to toe, with nothing cropped outside the frame. The camera is positioned directly in front of her at eye level, perfectly parallel to the ground, maintaining a straight central frontal perspective with natural body proportions and no exaggerated perspective distortion. The camera is placed far enough away to comfortably frame her entire body while preserving realistic geometry and balanced composition.
She has naturally harmonious East Asian facial features with a softly balanced oval face, a delicate smooth jawline, gently defined cheekbones, and a refined feminine appearance. Her complexion is clear light beige with soft warm golden undertones, naturally healthy skin, realistic pores, subtle skin texture, and gentle facial color variation. Her long silky dark brown hair falls naturally over her shoulders with a healthy shine and soft movement.
Her medium almond-shaped dark brown eyes feature a subtle epicanthic fold and a calm, intelligent expression. She looks naturally toward the camera with warm, relaxed eye contact that feels genuine rather than posed. Her eyelashes are naturally long and delicate, while softly arched eyebrows create elegant facial harmony. She has a refined straight nose with a delicate bridge and softly rounded tip. Her naturally full lips have a soft pink tone and form a gentle relaxed smile that expresses quiet confidence, kindness, and emotional warmth.
She has a graceful, anatomically realistic feminine body with naturally balanced proportions. Her physique is slim and athletic without appearing overly thin. Her shoulders are soft and feminine, her waist is gently defined, and her hips create elegant natural curves. Her long legs are proportionate to her height, with graceful knees, slender calves, delicate ankles, and naturally shaped feet. Her arms are slim and lightly toned, ending in elegant hands with long relaxed fingers. Her neck is long and refined, complementing her balanced posture.
Her body language appears completely natural and relaxed. She stands comfortably with most of her weight resting on one leg, creating a subtle contrapposto stance with a slight shift of the hips and relaxed shoulders. Her arms rest naturally beside her body with soft, effortless positioning. She appears calm, approachable, confident, and completely comfortable within the environment, avoiding any mannequin-like stiffness.
She wears a premium casual outfit consisting of a cropped tweed jacket with elegant gold buttons over a fitted gray t-shirt, paired with high-rise dark indigo jeans that naturally follow the contours of her body. She wears clean white leather sneakers with subtle stitching, realistic material texture, soft creases, neatly tied laces, and naturally proportioned soles. Her footwear is fully visible within the frame.
The scene is set on a gently rolling green hillside beside the weathered ruins of a dramatic Gothic castle with broken towers, aged stone walls, and hollow pointed windows. Massive snow-covered mountains rise in the far distance beneath heavy storm clouds. Soft natural daylight filters through the clouds, creating realistic illumination, gentle contrast, soft shadows, atmospheric depth, and rich environmental detail. The composition remains spacious, symmetrical, balanced, and intentionally frames both the woman and the surrounding landscape without cropping any part of her body.
For you example
https://civitai.red/posts/30069634
State exactly what you mean...
Important update: in this version 2, I uploaded the INT8 and INT8-CONVROT quantized models, as well as the FP8 version.
Hi, the description says the checkpoint file should be named krea2-SAT-DirtyRealismV2-int8.safetensors, but the downloaded file is named krea2SATDirtyrealism_krea2SAT.safetensors. Are these the same checkpoint, or was the wrong file uploaded?
@firstloubikini4933 Hi, I'm not sure if this is a bug on Civitai's end or if it was renamed for some reason; the file was uploaded correctly with the name INT8.
Additionally, this is the one INT8-Convrot model I can't get to work no matter what
I even tried a normal diffusion loader, and it says it's 'INT8'
When I use the actual INT8 diffusion loader I get pure RGB noise.
Thank you very much. Sometimes people blame the model when something goes wrong without taking the time to really test it; receiving feedback from people who actually test the model—and many others—is what makes continuing to create models worthwhile.
Not only does it make for great realism, this knows editorial comic artists really well - somehow even better than base KREA for a few names. Just in case there's other fans of Jack Cole, Bill Ward, and Chon Day out here.
v2 is a huge improvement and looks awesome but the images are extremely over-sharp to the point many of them have spots or blotches from oversharpening. Check your monitor settings, bro maybe you have it configured so you don't see it but it's very obvious if you compare to all others.
there's a kind of "deinterlacing" effect, especially on hair. How can I prevent this?
Hi, I hope you could share an example so we can see exactly what you're referring to and how we might improve it.
@3216541781 use sampler:ER SDE, Res Multistep , NO use Euler
This model produces very weird "invisible mesh" artifacts on Forge Neo, even with the recommended setting.
Hi, I use Forge Neo for most generations. It would be helpful if you could share any of the prompts that are failing; in any case, I'm working on a version 3 that fixes those issues, especially at high resolutions.
Is it possible to download fp16/bf16 of your model somewhere?
Looks illegal omg I got framed real... Ouch.
This is such a weird checkpoint. It adds rain or wetness to 95% of generated pictures even with no mentions of it in prompts. Never seen anything like it.
Hi, all of that was fixed in version 3. Regards.