AI generated fetish images

[Like Ra]

If you want to test DALL-E 2 - use this Discord invite: https://discord.gg/7TJ5vx2GA3

Basic Instructions:
- Join a #newbie channel
- Type /imagine and then whatever you want
- The bot will send you 4 images in 60 seconds
- Click the numbered buttons underneath to get upscales (U) or variations (V)

You will have a limited amount of creations, then you have to subscribe for a paid plan.

[Like Ra]

This version of image generator is far from perfect, but it's really amazing what it can generate (I can't stop browsing the discord with all the generated and being generated images). It takes multiple attempts and time to get what you ask, so, I hope someone will create a free access  (e.g. with an ad-based business model).

This is what I could get for "a girl in a latex leotard with "likera.com" on it" (and an anime version of it):

       

Not much 😬

Not gonna lie this discord AI is pretty good.

The other Dall-E AI is also good. Though a bit slow and I can't get it to share images via link either.

[Like Ra]

I can't get it to share images via link either.

Please upload images into the forum, do no hotlink them.

Oh damn I've been using links this entire time. My apologies.
   

[Like Ra]

This is what I could get for "a girl in a latex leotard with "likera.com" on it" (and an anime version of it):

DALL-E v1 (in comparison to v2) does not understand text at all:

       

(I still like the images, though).

[Like Ra]

There is even a tutorial how to prompt the model correctly: https://dallery.gallery/the-dalle-2-prompt-book/

If you want to dig a little deeper, there is now a free standalone version of another AI generator, called "Stable diffusion" you can run independently on your own PC.

It is VERY impressive, or at least it can be, if you experimented with it a bit and know what you're doing, see my first tries below. Unfortunately, it does not seem to be trained to be very ... kinky, yet.
  
Here's the blog post announcing the release with some examples.

Here's an aggregator for work created with the online version and the associated text prompts.
It's insane what this AI is capable of if fed the correct prompts and given potent hardware.

Here is a step-by-step tutorial on how to set it up in complete layman's terms.

Source: https://www.youtube.com/watch?v=ZCJpIDADzTI


Everything is explained start to finish, I just completed the process myself and I have no real programming experience beyond html/css.

I didn't see any requirements mentioned, but you will probably need an nVidia 20x0 series or higher with CUDA cores and at least 8 GB of VRAM (or you could get out of memory errors). My 2070S was working up quite a sweat and with 8 GB I could not do 1024 x 768 in size, so I stuck to 512 x 512.

A 24 GB 4090 looks quite tempting now.

There are some bumps in the tutorial video, I am outlining them here so you don't run into them as well.

You will run into an error. He corrects his file structure at around 9:30 after Anaconda throws the same error for him, just follow along and it will work - until you get another error because the video is also missing the crucial step to install Git (he forgot to uninstall it, so he will not get an error, but you will if you follow along don't have Git set up), a follow-up video describes how to:

Source: https://www.youtube.com/watch?v=lt9oDAvpG4I


Do this before you build the environment in Anaconda with the command
conda env create -f environment.yaml
or you will get another error and will have to fix it by following the second video completely (you will have to remove the environment).

My first prompt was quite simple, no idea why the AI struggled so hard with it 😁
Nudity does not seem to be an issue though.

Code:

buxom burlesque dancer in a rope hogtie, ballgag in mouth

And the results, not quite unexpectedly, will keep you awake at night:

       
       

Code:

Blonde fetish model in red latex dress

   

Code:

dita van teese in a red latex dress

       
       


Getting better ...

Code:

close up of a beautiful female face in fashion makeup with a red ballgag in her mouth

   


And of course it had to be done, a random Rule 34 prompt for good measure:

Code:

"Full Body digital artwork of Mercy Overwatch, topless and wearing black latex stockings and a garter belt, wearing red leather collar, wearing makeup, ultrarealistic

       
       

All in all, pretty impressive results in a couple of minutes of tinkering, without needing an online service or subscription.

It certainly looks like no one will have to hire someone to design an album cover ever again.