<h3 id="my-advice-which-one-for-what-purpose">My advice, which one for what purpose?</h3><ul><li>Graphic, illustration &amp; painting : V2, V3 &amp; V6</li><li>Photorealism : V1, V4, V5, V6 &amp; V7</li></ul>You can also use <a target="_blank" rel="ugc" href="https://civarchive.com/models/2838525/kreation">Kreation </a>for drawing, graphic design, illustration and painting.<h3 id="variant-7">Variant 7</h3><ul><li>Exclusive focus on photorealism</li><li>Improved lighting and texture rendering, particularly for skin (blood vessels visible through the skin)</li><li>Attempt to expand NSFW rendering capabilities (following persistent requests)</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3a26f17d-2f58-42dd-b793-9d17060ff1b8/original=true/3a26f17d-2f58-42dd-b793-9d17060ff1b8.jpeg" />Note : I’ve spotted an artifact that appears completely at random (the kind of thing that drives you crazy), making it impossible to pin down. I’ll let you guess which one it is. If you happen to notice it in a render, simply run a different seed to make it disappear. <h3 id="variant-6">Variant 6</h3>I'm currently only distributing the optimized inference formats (INT8/ConvRot version) to keep performance high and protect the core workflow. The BF16 master weights won't be released publicly for this version. Thanks for understanding!The V6 isn't a major overhaul, it's just a series of adjustments to the V5.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/21c77dfb-0593-4ada-b864-6021bde1a342/original=true/21c77dfb-0593-4ada-b864-6021bde1a342.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f1143dd3-145c-483a-a094-bc354e373104/original=true/f1143dd3-145c-483a-a094-bc354e373104.jpeg" /><h3 id="variant-5">Variant 5</h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/159cc837-8fae-48c6-b8d9-89685393fd4c/original=true/159cc837-8fae-48c6-b8d9-89685393fd4c.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b703118a-719a-47f6-9d5d-26a297b92ee7/original=true/b703118a-719a-47f6-9d5d-26a297b92ee7.jpeg" /><h3 id="more-face-details">More face details</h3>For those wanting even more detail on close-up portraits and tight face framing,using FaceDetailer (Ultralytics face and eye bbox) is, of course, essential.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/382df8dd-d22b-46a0-9dd7-23dc420c6dc2/original=true/382df8dd-d22b-46a0-9dd7-23dc420c6dc2.jpeg" /><h3 id="my-settings">My settings</h3>For achieving natural-looking skin renders with V4.first pass : 10 steps / rk6_7s / linear quadratic / denoise 1 second pass : 4 steps / abnorsett_4m / bong_tangent / denoise 0.2<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/302e3c0d-ac33-41e8-835c-45446f2ff466/original=true/302e3c0d-ac33-41e8-835c-45446f2ff466.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/42dc5f02-cbad-45ae-aaab-d25a81fa5133/original=true/42dc5f02-cbad-45ae-aaab-d25a81fa5133.jpeg" /><h3 id="variant-4">Variant 4</h3>There's no point in chasing two hares at once, Variant 4 is oriented exclusively towards the photorealism, The pictorial aspect will be considered exclusively in subsequent versions..<ul><li>V4 has been completely uncensored, if any censorship persists, it stems from the Qwen3VL text encoder, though this remains very rare.</li><li>The colors have been kept to a very neutral and natural palette.</li><li>GGUF Q8 &amp; Q4 <a target="_blank" rel="ugc" href="https://civarchive.com/models/2774280/kreamania-v4-gguf?modelVersionId=3123839">https://civarchive.com/models/2774280/kreamania-v4-gguf?modelVersionId=3123839</a></li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/24560d8b-ef25-4b01-91ba-22a1b430275f/original=true/24560d8b-ef25-4b01-91ba-22a1b430275f.jpeg" /><h3 id="variant-3">Variant 3</h3><ul><li>Improved skin texture for greater realism</li><li>Optimized lighting effects, more subtle and with better contrast</li><li>Enhanced painterly visual style</li><li>Corrected color shift toward warm tones, color rendering is more neutral and realistic</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6e111a03-c7c8-4533-afb3-3eff08c76aac/original=true/6e111a03-c7c8-4533-afb3-3eff08c76aac.jpeg" />Feedback on V3<ul><li>V3 is a lot more restrictive</li></ul>Yes, the "unlocked" feel has been diluted; I’ll need to revisit that in future versions.<ul><li>Colors are less realistic.</li></ul>Yes, that aspect still needs fine-tuning in subsequent versions.<ul><li>Less realistic than V1</li></ul>That’s possible, given that I’m aiming to finalize a versatile model, one that is realistic for photography yet also graphic in style.KREA 2 TURBO Vs KREAMANIA<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/01a2fa8b-d617-4c6a-91ff-5fb5998e25b9/original=true/01a2fa8b-d617-4c6a-91ff-5fb5998e25b9.jpeg" /><h3 id="variant-2">Variant 2</h3>Improvement of version 1 by merging an unpublished personal LoRAs to optimize lighting, textures, and graphic effects.<h3 id="variant-1">Variant 1</h3>The removal of certain restrictions compared to the model turbo fp8 scaled, along with improved realism and nudity rendering. The goal wasn't to focus on an NSFW model (even though it is capable of that), but rather to unlock the full range of creative possibilities. Recommended ComfyUI extensions:<a target="_blank" rel="ugc" href="https://github.com/ethanfel/ComfyUI-Krea2TextEncoder">https://github.com/ethanfel/ComfyUI-Krea2TextEncoder</a><a target="_blank" rel="ugc" href="https://github.com/capitan01R/ComfyUI-Krea2T-Enhancer">https://github.com/capitan01R/ComfyUI-Krea2T-Enhancer</a><a target="_blank" rel="ugc" href="https://github.com/huwhitememes/comfyui-krea2-conditioning">https://github.com/huwhitememes/comfyui-krea2-conditioning</a>

Kreamania

New approach of V7- Complete abandonment of artistic rendering styles (such as graphic, painterly, or illustrative looks); the model focuses entirely on realism.- The exclusive focus on photorealism has improved the balance of lighting and textures. Skin rendering achieves a high level of quality, regardless of the framing distance.- Repeated requests to open the model up to NSFW content—specifically sexual scenes—presented a challenge, as maintaining high rendering quality is difficult in this context. I decided to give it a try anyway, and I believe this variant performs best in this area, without being classified as an exclusively NSFW-oriented model.The trigger is not essential.

Files with SHA256: 29c049c48838bf547c8388eb86aa350b647918e755b3f3df9eef8e36bb831162