20s on 20 - Animagine XL v3

The 20s on 20 Project

This is a comparison of images made by twenty different AI models, all using the same prompt... one featuring a 1920s model. Different prompts work better on different models, so experiment.

After using all these models for this project, and also from trying it on other prompts before, I can say that Animagine XL v3 is the worst of the XL AI models. Absolute bottom barrel. Not only is anime not great to begin with, but this is the worst of the models using that style. It's like they dug up an animator from the 1960s who just had one image for like 5 minutes and would just animate the mouth moving. Maybe people are used to that art style with visual novel videogames. They chose that style because it's cheap art to produce and requires little skill. If you can choose any other style, why choose this one?

If you want to see all the comparison images, try "the-20s-on-20" tag.

/tag/the-20s-on-20