no NSFW for now thats the catch, realistically speaking though unless u have like lots of VRAM itll take forever to render 10s vid
Photos & Videos AI Tools: How to and Technical Discussion on Creating NSFW Gay Sex Stuff with Stable Diffusion (and others)
- Thread starter StudmeatFL
- Start date
-
- Tags
- bara civitai men stable diffusion
Does this no longer let us create NSFW stuff?
https://tensor.art/models/760875391155065513/VirileStallion-5.0-5.0
https://tensor.art/models/760875391155065513/VirileStallion-5.0-5.0
Still worked for me
Prompt
naked muscular male kneeling outdoors by a swimming pool, looking over right shoulder directly at viewer, right hand on ass, view from side, back arched, legs spread wide, short black buzzed haircut, 40yo, mouth open in ecstasy, huge uncut erect penis, covered in cum, very sweaty, full body view, masturbating, ejaculating, hand on penisNegative prompt
EasyNegative, shower fixtures, shower hardware, SFWSteps
25Sampler
DPM++ 2M SDE KarrasKSampler
dpmpp_2m_sde_gpuSchedule
karrasCFG scale
7Guidance
3.5Seed
146593007Size
768x1152VAE
AutomaticDenoising strength
0Clip skip
2Model
virileStallion_v50PhotorealLoRA
extreme_bukkake_v0.1-pony:0.80
Tensor.art is great because you can start for free.
I did a walk through of their interface here:
Post in thread 'AI Tools: How to and Technical Discussion on Creating NSFW Gay Sex Stuff with Stable Diffusion (and others)'
Photos & Videos - AI Tools: How to and Technical Discussion on Creating NSFW Gay Sex Stuff with Stable Diffusion (and others)
Guess I kinda successfully create it using the airfuck's brute mix v1.0 with face editor. Thanks so much tor the tutorial.
![image (1).png]( "image (1).png")
yeah go rent a cloud service - u can run it on the ur web browser
reduce the cfg scale?I haven't found any discussions on upscalers... I used to have a nice one on my old machine. I thought it'd be easy to retrace my steps but...So, I'm stuck with stuff that makes my guys look like they're made of plastic.
Reducing the cfg gives me good results when I'm not wanting to get all creative, and want to make the prompt be fully followed. I generally go from 6ish to 7ish, on that... But I've played with 2, 3, 18... Anyway... The issue right now is how the first images I originally produce have better, more detailed skin than the upscaled version afterwards. I've been battling with different upscalers, but it is a world out there, and not just connected to AI images... My website of reference for these issues is openmodeldb.info but there are just too many different upcalers, for too many situations... Sometimes, some grain in the image produces more realistic stuff than these super clear ones that are on the rise,... Anyway, thanks for the input. I'll keep trying.