Finally, how is my art made with AI?

Welcome to the creative process of my artwork on PixAI!

In this demo, I will show you how I create my artwork using artificial intelligence with the help of the PixAI platform. This is a small glimpse of what I share with my premium supporters, where I will show you the steps I follow to generate the artwork and how I make adjustments to the final result.

After registering on the platform, you will receive credits, and with them you can create artwork. You can collect credits every day, especially if you download the app on your phone. But the app is very limited, so we'll stick with the website. When you create artwork, your platform will look something like this:

Here you will create your art with prompts and tools that the platform offers you. It is worth remembering that there are a lot of tests that will need to be done and each one has its cost. Many times it will not turn out well and it will be necessary to make some adjustments.

But let me give you an example of an art I made

1. Choosing the Prompt

The first step is to define what I want to create.

For this artwork, I used the following prompt as a base:

"woman flexing her biceps with a pose in a gym room, 1girl, solo, blonde hair, long hair, realistic, looking at viewer, sports bra, breasts, muscular,,hannahowo,SlightMuscle, Bulky, Bodybuilder, Hypermuscle,muscular female"

This prompt allows me to create a strong and confident character, which is one of my favorite themes. With that, the next step is to add reference images.

Remember, this is an example where I used a reference image, I often don't even do this, but for those who are starting out it can be a great help to start getting it right more.

So I went to look for an image to base myself on and extract a prompt from it... yes, the platform can help you with the prompts

I found this random image on the internet...

In this part I can put the image found... Remember, be careful with copyrights!!!

We reduced its strength, to make it as close as possible to the original. This way we can generate it and have the identical image, use it as a base for the next image and also have its prompt.

The tool can help you with some prompts about the created image, this way you can have less work to create a prompt from scratch and work on that basis.

And now the testing begins... it depends a lot on the model that will be used and the Loras that will be added what will be generated in the end.

This will take time, patience, adjustment and frustration. But some are worth it!

Pulling towards reality, depending on the model

Or more towards 3D realistic

Testing lora is what I do most, that's why my art sometimes comes out quite varied, but it's all a matter of testing, testing and testing...