After testing over 4000 different PDXL, ILXL and NoobAI models, I decided to test Animagine XL 4.0 based models as well. The general idea stays the same: I only want to give a comparison of their default styles without judging them. Animagine XL 4.0 has a later knowledge cut-off compared to PDXL, ILXL and NoobAI (January 7th 2025), so it already knows newer characters. Although Animagine also has a limited understanding of natural language, I mostly kept the same prompt as with ILXL/NoobAI and only adjusted the quality/score tags according to Animagine's prompt structure.
Just like with PDXL, ILXL and NoobAI I will only use SFW prompts for my test images and censor images that go against CivitAI's TOS. But keep in mind that Animagine is perfectly capable of producing NSFW images as well.
This comparison is not completely fair for every model, but adjusting the prompts and looking for a good seed for every model would not provide a good comparison. I download almost every new model and post the test images to their respective model pages with my test image upload account Lizardon1025.
All the prompts and settings are listed below. You can also find my updated ComfyUI workflow for all test images attached to this article.
Since Animagine does not have its own base model identifier on CivitAI (yet), I might miss some new Animagine XL 4.0 based models. But you can always just contact me via CivitAI chat and let me know about it.
A list of my personal favorite Animagine models (like I did with PDXL and ILXL/NoobAI) might follow in the future.
The test prompts are the following:
Zelda kiss:
1girl, 1boy, princess zelda, the legend of zelda: breath of the wild, standing, kiss, hug, closed eyes, outdoors , masterpiece, high score, great score, absurdres
Portrait:
1girl, face, portrait, looking at viewer, closed mouth , masterpiece, high score, great score, absurdres
Forest Angel:
1girl, solo, blonde hair, wavy hair, angel, angel wings, halo, smile, sitting, forest, white dress , masterpiece, high score, great score, absurdres
Library Catgirl:
1girl, solo, long hair, black hair, blunt bangs, straight hair, round eyewear, cat girl, cat ears, cat tail, standing, hug, cat, looking at viewer, library , masterpiece, high score, great score, absurdres
Negative prompts:
lowres, worst quality, low quality, low score, bad score
All images were created with their recommended sampler, scheduler and CFG value. If there are no recommended settings to be found, I'm using these default settings:
Sampler: Euler a
Scheduler: Normal
CFG Scale: 5
Steps: 100
Width: 896
Height: 1152
Seed: fixed