Photos & Videos AI Tools: How to and Technical Discussion on Creating NSFW Gay Sex Stuff with Stable Diffusion (and others)

[Fjsuhwr]

Well, I did play around with Tensor.Art's Training for Lora and it gets the face and body really close if you use Flux; unfortunately, I'm not finding any Flux based Checkpoints for NSFW stuff. I did a train with Pony and it looked like utter crap. It's probably more me not knowing what I'm doing.

Did u train offline or online? There r good tools offline that can be used. If Lora doesn’t work then try training a separate whole model

 

[VanMetz]

Just starting out with generation and trying to improve. Wondered if anyone has any advice or tips on how to take my images to the next level. I'll post a few here as examples...

Nice

 

[ThePariah]

Any suggestions for an AI tool that let's you in-paint a cock into a painted-style image? I have some great ones I've made with Sora that I'd love to in-paint in.

 

[StudmeatFL]

Did u train offline or online? There r good tools offline that can be used. If Lora doesn’t work then try training a separate whole model

I don't have the necessary computer equipment to do the offline training

I mean I do have an M4 Macbook Air, but I've seen what it does in Diffusion and it's pretty much 'nope'.

 

[StudmeatFL]

Any suggestions for an AI tool that let's you in-paint a cock into a painted-style image? I have some great ones I've made with Sora that I'd love to in-paint in.

There is an AIDetailer specific for cocks and balls:
Cock and Ball Detection 2D edition (ADetailer) - v2.0 | Other Detection | Civitai

 

[StudmeatFL]

There is an AIDetailer specific for cocks and balls:
Cock and Ball Detection 2D edition (ADetailer) - v2.0 | Other Detection | Civitai

Whoops, that one is only 2D, here's the Penis detailer that works with AIDetailer:

Penis + Pussy Adetailer Model - penis | Stable Diffusion Other | Civitai

 

[Fjsuhwr]

I don't have the necessary computer equipment to do the offline training

I mean I do have an M4 Macbook Air, but I've seen what it does in Diffusion and it's pretty much 'nope'.

aha thats tough

 

[StudmeatFL]

aha thats tough

I've had good luck with Google Collab in the past, just need to do one of those again with a paid device. At present, just playing with Tensor.Art randomly.

 

[robert29201]

literally this lol

heres for the salty poster thinking that this thread is for credits - if u have a powerful enough rig, have fun running it locally offline NO CREDIT needed:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

That's what I did. I have a computer that I hadn't turned on in a year. AMD Ryzen 9 processor with an Nvidia RTX something video card and 96gbs of memory. Stable Diffusion and Flux fly on that thing. It's amazing and fun to generate generate generate and not have to pay. LOL

 

[Kayrizz]

Well, I did play around with Tensor.Art's Training for Lora and it gets the face and body really close if you use Flux; unfortunately, I'm not finding any Flux based Checkpoints for NSFW stuff. I did a train with Pony and it looked like utter crap. It's probably more me not knowing what I'm doing.

Valid, i also dont know what im doing LOL

 

[ThePariah]

Any suggestions for apps that allow NSFW content where you can add an audio file to a video file AND sync the mouth movement to the audio?

I have some great AI vids that I'd love to add audio to, but I'm guessing most sites won't accept NSFW.

 

[StudmeatFL]

Any suggestions for apps that allow NSFW content where you can add an audio file to a video file AND sync the mouth movement to the audio?

I have some great AI vids that I'd love to add audio to, but I'm guessing most sites won't accept NSFW.

No clue, but following to see if anyone has any ideas

 

[Tunsoffun]

I don't have any experience with a photo editor like photoshop. Is there an AI I can use to give this guy a worthy bulge?

 

[xyz45456]

I don't have any experience with a photo editor like photoshop. Is there an AI I can use to give this guy a worthy bulge?View attachment 175874691

Dunno about online resources but SD's InPainting function does it 'problemlos' (see attachments).

 

[Tunsoffun]

Dunno about online resources but SD's InPainting function does it 'problemlos' (see attachments).View attachment 175942011

Awesome. What prompt did you use?

 

[xyz45456]

Awesome. What prompt did you use?

Just selected the area and used the InPainting default tensors (sd-v1-5-inpainting.ckpt) with "bulging" or "bulging shorts" as prompt. The "whole picture" option - not "only masked" - then delivered it--only a slight colorimetric difference.

 

[Tunsoffun]

Just selected the area and used the InPainting default tensors (sd-v1-5-inpainting.ckpt) with "bulging" or "bulging shorts" as prompt. The "whole picture" option - not "only masked" - then delivered it--only a slight colorimetric difference.

Are you doing this locally? What UI are you using? I tried it on my local Automatic1111 with Juggernaut and it failed miserably.

 

[xyz45456]

Are you doing this locally? What UI are you using? I tried it on my local Automatic1111 with Juggernaut and it failed miserably.

Yes, locally. StableDiffusion 1.5--the old but highly reliable one. Sorry to hear it doesn't work with Automatic 111. Maybe there are version-specific InPainting tensors you could download...