I previously tried artsmart.ai and USP.ai. Although they have some features that Artsmart does not (like choosing different models) the quality of the images was pretty poor compared to something like Midjourney or Lexica.ai.

I actually used Lexica.ai to get some prompts in case I wasn't very good with prompt making, and put prompts into all three apps. Artsmart by far with their hyperrealistic model was the winner, creating images that were absolutely on par with the Lexica.ai outputs.

In addition they have some features other place don't like being able to erase part of the image and regenerate it (inpainting) with outpainting to be added soon. They also added a Hyperrealistic mode after I joined which is great to play with.

I also like that they have a lot of prompt suggestions and massaging, like each image has a default negative prompt that I think really helps the images out of the box vs the ones that don't add this.

I was able to make a small anime themed Visual Novel for a friend and was blown away at how usable the images were. On the other apps I had to regenerate again and again and again and got subpar images, but their Semirealistic model is just top notch.

They also have a "tunes" feature which allows you to create images based on say a photo of yourself or a friend which is very unique among the other apps I've seen.

Unlike the other AI lifetimes on here the creator seems receptive to feedback, responsive, and is actively adding features.

All in all buy this, you won't regret it!