Hi, I tried something...

This is my first attempt at an Anima-based checkpoint model, so I'm still very inexperienced with Anima, and it is very different from SDXL models, so I'm still not sure how it'll perform for other people.

For those who are new to Anima models, the prompting is pretty similar to Illustrious XL and NoobAI models. It also primarily uses Danbooru and Gelbooru-style tags. Artist style tags work similarly to ILXL/NAI models, but you have to add @ in front of the artist name. So instead of dizdoodz, it has to be @dizdoodz. Character tags work the same way with ILXL/NAI.

Tips and Recommended Usage

Feel free to experiment with different styles, character types, prompts, and LoRAs. Results may vary depending on your workflow and settings.

To use it, you need VAE and CLIP; recommended these:

VAE CLIP

Unless you're using Civitai's generator, you'll need either ComfyUI or ForgeNeo to run it. Check the official Anima model page [HF] for the required text encoder and VAE file that's necessary to run it.

Positive Prompt:

masterpiece, best quality, amazing, score_9, score_8,score_7, very aesthetic, extremely detailed, cinematic,

Negative Prompt:

worst quality, low quality, score_1, score_2, score_3, artist name, blurry, jpeg artifacts, lowres, censor, bad anatomy, watermark, artist name

Good Setting:

CFG: 4 / 6
Steps: 30 / 40
Sampler: Er_sde / Euler A
Scheduler: Simple
Layout: same that SDXL
Upscaler: R-ESRGAN 4x+ Anime6B
Hires fix: upscale - 1.5x
           upscaler - R-ESRGAN 4x+ Anime6B
           denoising - 3.5 - 4