Software I use to make AI Art

I give an overview of all the different software applications I used as well as AI Art websites and resources I use.

Software I use to make AI Art
Editing an image of Ellen Hartley in Affinity Photo 2

I get many questions about what program or service I use. What my workflow is. In my AI Art school I intend to go through that in detail, but let me just start with the question everyone asks: What do I use?

The very short answer is that I used Automatic1111. Automatic1111 is a front-end to Stable Diffusion that you run locally on your computer. Or, you can in theory run it in the cloud on services such as Hugging Space (more on that later). I used to do that earlier.

The longer answer is that I use lots of software and services. There is no tool or AI model that can do everything well. The images I am most happy with are often made using a combination of many tools, services and AI models. Exactly how I do that will not be covered here. Rather I will give you an overview of the tools I use or have used.

What do you want to make?

What you want to make decides a lot on what service you might want to pursue. If you would like to make SFW (suitable for work) content, your options are much richer and varied than if you want to make NSFW (not suitable for work) content. If you would like to make nice artistic non-pornographic images, then Midjourney, Leonardo AI and Lexica are probably the services I have had best experiences with.

But presumably you would not be getting advice from me if that is what you wanted. I make lots of naked girls with big tits and big butts. If you didn't care about that at all then you most likely wouldn't follow me or search for advice from me. So I assume you are going to want to make naughty images. In that case, your options are more limited, but that is what I will describe in the next section.

Mage Space is the first online service I used for NSFW (not suitable for work) images
Mage Space is the first online service I used for NSFW (not suitable for work) images

Naughty Images for Beginners

If you want to get started with AI Art and don't want to blow lots of money and you want to make some NSFW (not suitable for work) images then here are my top choices:

  • Mage Space — All my early work on Deviant Art of busty ladies was done on Mage Space. It has become incredibly versatile since began using it
  • PromptChan — Very focused on nudes and pornographic content.

I only have very limited experience with PromptChan, but I have tried it briefly myself and seen what others makes so I think it is safe to say that it is a great site for making sexy nudes or even images involving sexual intercourse. You can even create NSFW videos.

Mage Space, however, I have used extensively myself and know many other you have used it with success. It is far more versatile than PromptChan, which has a singular focus. Only criticism I have of Mage Space today is that it has perhaps gotten a bit too complex. I never thought user interface was their strong suit.

Non-Nude Beginner Advice

If you aren't making nude busty girls then your choice of service is far wider. These would be my top choices:

  1. Midjourney — Stunning images and easy to use. Great artistic flair and variation.
  2. Lexica — Very easy to use. Crisp images. Very good for robots and mechanical things.
  3. Leonardo AI — Rich set of tools but very well designed interface.
  4. Replicate — Cheap way to run Flux AI model

This is roughly in my order of preferences. I am a huge Midjourney fan. It makes stunning images very quickly and easily. There are other tools which make good looking images but they tend to lack the artistic flair or sheer variation Midjourney offer. With Midjourney you get out images with huge variation in style and look. If you want characters with strong expressions, dramatic light, or something that looks like a movie scene with a lot of action you cannot beat Midjourney.

Midjourney is one of my favorite online AI Art generators
Midjourney is one of my favorite online AI Art generators

All AI Art generators however tends to have a certain style. Once you use a lot of them you start getting a feel for this style. Lexica for instance has a sort of crispness to its style that somehow lends itself very well to making robots and mechanical devices. 

Lexica AI Art generator website
Lexica AI Art generator website

While Midjourney and Lexica have built their own unique AI Art generators, many of the other services are more like interfaces to well known AI Art generators such as Stable Diffusion or Flux. With Leonardo AI or Mage Space, for instance, you can load up huge numbers of different AI models tailored for making different style images. It may seem like these services would then be much better than Midjourney or Lexica but that would be a mistake to think.

The business model of these companies and services are entirely different. Leonardo AI and Mage Space don't really focus on building AI models. Rather they focus on creating good interface to AI models made by others.

Midjourney in contrast built their whole service around creating a high quality model. Their users have been part of making their output such high quality. Typically with Midjourney you rate images a lot. They will incentivize this in different ways. With a large user base rating images they have managed to tune their models to give spectacular results. It is almost impossible to create ugly images with Midjourney.

I could go on talking a lot more about these different services and their strengths but if you are interested in learning more of that you should follow my AI Art school as I will dig deeper into these topics.

My AI Art Tools and Services

To make my art I use software such as Automatic1111, Forge, Topaz Gigapixel AI and Affinity Photo 2. But I also use regularly services such as Midjourney, Kling and Replicate. Let me cover each of them to give an idea of what I use them for.