CinEro-NG-Krea2

Creator:
 <a href="https://civitai.red/user/homoludens" rel="noopener noreferrer nofollow" target="_blank">
 https://civitai.red/user/homoludens
 </a>




 
 About this version
 

<h3>
 v3e
</h3>

 TL;DR; Main difference from v3-alpha is better ethnicity adherence and face diversity; reduced dithering and increased texture / hairs smoothness.


 Introduction of trigger words means better adherence to the ethnicity in prompt. v3-alpha has weaker distinction between those nationalities. Also, the secondary goal of the training was the improvement of textures and reduction of dithering / regular grid pattern issue.
 
 Picture must be little bit smoother in v3e.

<hr/>

 This is the CinEro NG Krea 2 series of checkpoints. Aimed to deliver the dark cinematic SFW / NSFW picture with unrestricted Qwen 3 VL Heretic Text Encoder. Trained with sci-fi horror themed moody portraits dataset. Developed mostly to be used in ComfyUI, but should work in other Stable Diffusion generation software.


 This series of the models is focused on moody, atmospheric, creative portraits and picturesque landscapes.

v3e - int8

Creator:
 <a href="https://civitai.red/user/homoludens" rel="noopener noreferrer nofollow" target="_blank">
 https://civitai.red/user/homoludens
 </a>




 
 About this version
 

<h3>
 
 v3-alpha
 
</h3>
<ul>
 <li>
 
 More training on sci-fi, horror, street photography, cinematic portraits (sfw/nsfw).
 
 </li>
</ul>
<ul>
 <li>
 
 Not a final quality. Probably can give dithering / grid patterns in some cases for Text-to-Image pass. Also, ethnicity difference is not so high.
 
 </li>
</ul>
<hr/>

 This is the CinEro NG Krea 2 series of checkpoints. Aimed to deliver the dark cinematic SFW / NSFW picture with unrestricted Qwen 3 VL Heretic Text Encoder. Trained with sci-fi horror themed moody portraits dataset. Developed mostly to be used in ComfyUI, but should work in other Stable Diffusion generation software.


 This series of the models is focused on moody, atmospheric, creative portraits and picturesque landscapes.

v3-alpha

Creator:
 <a href="https://civitai.red/user/homoludens" rel="noopener noreferrer nofollow" target="_blank">
 https://civitai.red/user/homoludens
 </a>




 
 About this version
 


 Little bit of additional fine-tuning.

<hr/>

 This is the CinEro NG Krea 2 series of checkpoints. Aimed to deliver the dark cinematic SFW / NSFW picture with unrestricted Qwen 3 VL Heretic Text Encoder. Trained with sci-fi horror themed moody portraits dataset. Developed mostly to be used in ComfyUI, but should work in other Stable Diffusion generation software.


 This series of the models is focused on moody, atmospheric, creative portraits and picturesque landscapes.

v2i-6steps

v2g - Turbo

v2b - Turbo

Creator:
 <a href="https://civitai.red/user/homoludens" rel="noopener noreferrer nofollow" target="_blank">
 https://civitai.red/user/homoludens
 </a>




 
 About this version
 


 Made from v2a-turbo-BF16 by quantization of all tensors to FP8

<hr/>

 This is the CinEro NG Krea 2 series of checkpoints. Aimed to deliver the dark cinematic SFW / NSFW picture with unrestricted Qwen 3 VL Heretic Text Encoder. Trained with sci-fi horror themed moody portraits dataset. Developed mostly to be used in ComfyUI, but should work in other Stable Diffusion generation software.


 This series of the models is focused on moody, atmospheric, creative portraits and picturesque landscapes.