Pics AiHunks - Ai Porn <Stable Diffusion> Photos, Opinions (artifical intelligence) Machine Learning NSFW

[AiHunks]

heres a few more from the community of unstable diffusion:

 

[72-Jay]

.. Note the 'old guy' in the background (who's totally staring at this dude)...
Both him and the person beside have mis-matched shoes! that's a bug I've noticed in the Stable Diffusion AI.

-----
Its amazing just how convincingly-real these AI systems like Stable Diffusion can make an image look (not just people, but nature-scenery, buildings, objects, etc)
Also I found in Stable you can also do stuff like make a guy look sad or crying.

The site I use also has 2 others:
'CLIP-Guided Diffusion' and 'VQGAN+CLIP'

 

[AiHunks]

recently got a new computer and was able to generate some animated .gif Aihunks
heres a few from my spartan series





still lots of room for progression with these animation Ai models.
but happy to start using them.
what do you think? will Ai generated movies be a thing one day?

 

[InsanePath]

recently got a new computer and was able to generate some animated .gif Aihunks
heres a few from my spartan series

View attachment 1873141
View attachment 1873142
View attachment 1873144

still lots of room for progression with these animation Ai models.
but happy to start using them.
what do you think? will Ai generated movies be a thing one day?

Stuff is amazing

 

[72-Jay]

@AiHunks:
Cool on the new computer!
Just a matter of time before someone does an AI movie.
I've tried couple little 'videos' with the 'VQGAN+CLIP' AI on that site I use. neat stuff

The one I'm deff looking forward to is when a true 'AI for audio' comes out (that's far more complex than images or video, so I don't expect to see it for awhile)

 

[AiHunks]

Here is some images from the unstable diffusion community

 

[Paws]

This guy does not exist.

 

[OldLee1971]

I just got into tinkering with the AI art generation, and don't quite have a handle on the prompts... Does anyone know where I can find some decent examples? Especially for accurate anatomy? The one's I'm using do not create anything remotely "human." LOL

Below is an example of what I'm typically getting.

 

[AiHunks]

I just got into tinkering with the AI art generation, and don't quite have a handle on the prompts... Does anyone know where I can find some decent examples? Especially for accurate anatomy? The one's I'm using do not create anything remotely "human." LOL

Below is an example of what I'm typically getting.

yes i can assist with this.
could you provide me some more info on which Ai app youre using?

if you're using the A1111 stable diffusion install then i recommend using civitai models + textual embeddings.
currently im using the Unstable Diffusion Collaboration Model <- click

textual inversions im using:
BADDREAM textual inversion -<- click
Fast Negative Embeddings textual inversion <- click

I'm getting decent results , heres a few










what do you think?

 

[AiHunks]

Heres several from the unstable diffusion community

 

[AiHunks]

here's 10 more from the unstable diffusion community:

 

[AiHunks]

studio ghibli dick
made this earlier with batch img2img and control net posing + studio ghibli lora

 

[AiHunks]

heres some motion gif ive been making with AnimateDif:

 

[72-Jay]

AnimateDif:

Whats that? never heard of it. Does it just work with people? Or can you do something like a thunderstorm?

Sorry to say, but those animated ones are quite creepy. .lol. its the eyes!

 

[AiHunks]

Whats that? never heard of it. Does it just work with people? Or can you do something like a thunderstorm?

Sorry to say, but those animated ones are quite creepy. .lol. its the eyes!

for more info on the addon see link below
AnimateDif A1111 Extension program

yes you can do basically anything the Ai model has been trained on.
heres a little orange fox

AnimateDif an addon extension for an Ai program that makes animated .gif using Ai & motion templates.
unfortunately the faces are distorted because the Ai model im using SD 1.5 if only trained to produce 512x512 pixels images so when i stretch beyond that it doesnt look the best.



also have been using SD XL ai models to make higher resolution images ( 1024x1024 pixels)
eventually animation Ai options will be available for this higher resolution option.
heres some examples from SDXL :






SDXL is still fairly new and theres still a strong NSFW filter within the core model. so within time there will be more options for creating adult content with these new better moels.

another Text to video option aside from basic .gif include "runway ML"
GEN 2 Text to Video

 

[72-Jay]

AnimateDiff looks interesting. If I was running Stable Diffusion on my own computer I might try it.

I like the fox ... but it like the rest suffers from "creepy eyes"
Its actually not the stretching from default 512x512 that does it...its the fact each frame is generated independently & it doesn't track positioning of the eyes correctly between frames. I guess you could call it an 'ai bug' (probably something that will be worked out in future versions with more programming and/or training)

The clarity on those last few is pretty impressive!
The site where I use to generate AI images has a SDXL option, but I've never used it

 

[AiHunks]

heres a few SDXL images i made recently using the Xpenis lora

 

[AiHunks]

 

[maryjayne]

 

[cagedtwunk]

here's 10 more from the unstable diffusion community:

View attachment 2090469
View attachment 2090470
View attachment 2090471
View attachment 2090472
View attachment 2090473
View attachment 2090474
View attachment 2090475
View attachment 2090476
View attachment 2090477
View attachment 2090478

Are you actually able to share how you make these? I’ve been dabbling in AI creation but can’t seem to prompt anything nearly as realistic as these