In the world of ai generated images, three AI models stand out for their unique capabilities and potential to create images from a simple text prompt: Midjourney, Dall-E, and Stable Diffusion. By inputing a simple sentence it generates you what was described in it.

Dall-E

In June 2022 Dall-E has generated this image for me with this prompt: 'An oil painting of a fire contact juggling wizard'.

Pretty impressive, right? The image generation with Dall-E is resourceful and creative.

Midjourney

Midjourney is very user friendly. All you have to do is enter the discord chat, enter a chatroom and enter /imagine An oil painting of a fire contact juggling wizard and it gives you the created output. The latest version of stable diffusion up to this post is version 4 which was released in mid November 2022 and it is incredible! Six months after the first render with Dall-E, midjourney clearly puts Dall-E in it's shadow with it's incredibly detailed and accurate image generations. These two images were generated using the following prompt:

/imagine fire mage holds fire ball in his hand and juggles it around his body in an action pose in a dark forest, he has bright glowing blue eyes and long dark blonde hair high contrast, HDR, 8k, super detailed, --v 4

These were the first pictures that I generated with midjourney which really impressed me. But the guy doesn't look anything like the flow wizard. The flow wizard wears a hood and has a mysterious sharp look in his face. I wanted to get more of a portrait picture of a wizard which feels like the flow wizard to me which is why I tried a few variations of portrait prompts and came up with some pictures that I really started to like. Here I present you a couple of those results:

I really Love these images! Dark and mysterious. Here are a few more Midjourney generated images that impressed me very much: