BUZZ DONATIONS ARE HIGHLY APPRECIATED, SO I CAN BID THE MODEL INTO THE ONLINE GENERATOR

Feedback on v2.5 is greatly appreciated!

v2.5 is now available! This is an experimental release! Please provide feedback, it would be best to compare this version to the previous versions in your feedback and also mentions things that you like and dislike about it. Here are the improvements with 2.5:

• Improved skin details a tiny bit, you should notice more detail on faces and genitals. This should work without prompting for it, but you can also try the "detailed_skin" trigger word which was used during training for these details.

• Added film grain, without prompting for it there should be only a little bit (which helps make pictures look more natural), if you prompt for it with the "film_grain" trigger word you should get more pronounced film grain

• Tweaked facial features a bit to align with my own preferences, of course some people might not like these changes, please give some feedback if possible for this specific version :)!

v2.5 uses v2.0 as base, with 2 different loras that I made that were then merged into it. One was for the extra detail and the color and feature adjustments that I wanted, and the second one was for the film grain. v3.0 will use a bigger dataset and will be trained in 2 stages, so I don't have to merge my own loras, this should technically give better results, for now I need to see if I should make any adjustments before doing all of that though.

The point of this model is to have all the characters built into WAI-NSFW (full credit to WAI0731 for creating this incredible model), while having a more realistic look. I am hoping that in the near future I will be able to achieve near perfect photo realism.

!!! Important note !!!

Existing Loras won't work with this model since it is a fine tune, not a merge. It's possible that some loras might work, but most of them won't.

Popular loras that I have tested and confirmed that work well with my model: ExpressiveH, Detail Tweaker XL

ALL RECOMMENDATIONS BELLOW APPLY FOR BOTH V0.1 AND V1.0

Recommended settings:

Steps:: 15-30

CFG scale: 5-7

Sampler: Euler

Use size larger than 1024x1024 for the original dimensions

Recommended upscale settings:

Hires upscale: 1.5

Hires steps: 20

Denoising strength: 0.35~0.5

Hires upscaler: 2xNomosUni_span_multijpg_ldl (I tested a bunch of upscalers and this seems to work quite well for more realistic images, and it is pretty fast)

All the preview images have the workflow attached, it automatically calculates a 1.5x resolution for you so you don't have to change multiple fields each time you want to change resolution.

Everything is included in the checkpoint (vae, text encoders, unet)

Recommended prompts:

Positive Prompt

photorealistic, photograph, realistic, masterpiece, best quality, amazing quality, absurdres, detailed_skin

Negative Prompt

bad quality, worst quality, worst detail, sketch, censor, watermark, signature, text, multiple_poses, multiple_scenes, speech_bubble, patreon_username, multiple_images, multiple_angles, bad_hands, wrong_hand, bad_anatomy, extra_fingers, extra_digits 

Basically all characters in the WAI-NSFW model should work here too, you can check all the characters available here: https://huggingface.co/spaces/flagrantia/character_select_saa

Of course due to the nature of full fine tuning some characters might be impossible to be generated anymore, but from my testing all characters that I tried worked fine.

Old version update logs:

v2.0 is now available! This release delivers the following improvements:

• Slightly enhanced stability and prompt adherence across all outputs

• Expanded dataset with ~3x more training data (compared to v1.0), this helps diversify results

• 1536px training, yields more detail than previous versions

• Facial expressions have been improved, with previous versions smirks/slight smiles and other facial expressions often looked creepy, this is now mostly fixed, all facial expressions should look better (still possible to get bad results but a lot more rare now)

v1.0 is now available! This release delivers significant improvements:

• Enhanced stability and realism across all outputs

• Reduced CGI/plasticky look (still happens, but is more rare)

• Expanded dataset with ~2x more training data

• Multi-resolution training (1024px and 1280px vs 1024px only in v0.1)