After testing >8000 different Illustrious/NoobAI and >3000 different Pony Diffusion models, the release of Anima was a pleasant surprise. The Anima base model is already incredibly capable compared to Pony Diffusion, Illustrious and even NoobAI. Although the model file is smaller in comparison, Anima understands both Danbooru tags as well as natural language. It also has a later knowledge cut-off than former base models.
The general idea for this test stays the same: I only want to give a comparison of their default styles. Just as the Illustrious base model, Anima includes many different styles that can be triggered by using artist tags, so keep in mind that the default style (without artist tag) is not the only one the models are capable of. Just like with Illustrious, I expect those artist tag capabilities to diminish once fine-tunes of Anima push it towards a certain style. I kept the same prompt as with Illustrious but adjusted the quality tags accordingly.
The direct comparison of Anima with ILXL is a bit difficult since ILXL has way more finetunes and developed a great aesthetic look over time.
Just like with my former model tests, I stick to only use SFW prompts for my test images. But keep in mind that Anima 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. Although the recommended quality tags differ from model to model, I stick to the same set of positive and negative quality tags for all test images in order to improve comparability. You can find my ComfyUI workflow attached to this article.
Anima is still relatively new but I already have some favorite models This choice of models is entirely subjective. I will update the list from time to time.
Latest update of favorite list:
15.03.2026
Tested ILXL models so far: ~ 50
copycatAnima_20260302 (Euler a Simple, CFG ~ 6)
________________________________________________________________________________________________________
animaCatTower_v03 (Euler a Simple, CFG ~ 6)
________________________________________________________________________________________________________
animaCATANIMACitronAnime_v10 (Euler a Simple, CFG ~ 6)
________________________________________________________________________________________________________
animaRealAnimeCATAraCAT_v10 (Euler a Simple, CFG ~ 6)
The test prompts are the following:
Zelda kiss:
masterpiece, best quality, score_9, 1girl, 1boy, princess zelda, the legend of zelda: breath of the wild, standing, kiss, hug, closed eyes, outdoors
Portrait:
masterpiece, best quality, score_9, 1girl, face, portrait, looking at viewer, closed mouth
Forest Angel:
masterpiece, best quality, score_9, 1girl, solo, blonde hair, wavy hair, angel, angel wings, halo, smile, sitting, forest, white dress
Library Catgirl:
masterpiece, best quality, score_9, 1girl, solo, long hair, black hair, blunt bangs, straight hair, round eyewear, cat girl, cat ears, cat tail, standing, holding cat, cat, looking at viewer, library, dark, dim lighting
Negative prompts:
worst quality, low quality, score_1
I know that Anima can do natural language as well, which is a very useful feature, but I decided to stick to tags for my test images.
All images were created with their recommended sampler, scheduler, CFG value and steps. If there are no recommended settings on the model page, I'm using the settings of its showcase images. If the showcase images don't display their settings, I'm using these default settings:
Sampler: Euler a
Scheduler: Simple
CFG Scale: 5
Steps: 50
Width: 896
Height: 1152
Seed: fixed (same for all test images)