Redistribution is prohibited.

ABOUT THE MODEL

Jinhsi from Wuthering Waves

20260321v0.1

Trained on Anima Preivew 2

Anima can now easily achieve effects that I could never quite get right back in the SDXL era. The issue I’m running into, though, is that individual or even multiple tags don’t seem to bind strongly to the final output. No matter how I try to differentiate captions, I still end up with concept bleeding between them. I’m currently exploring possible solutions—though it might just be something the next version of Anima will address.

The current performance of Anima is already very well. However, when using latent upscaling beyond roughly 1.3× past the 1MP range, the results tend to break down, so I have to rely on alternative upscaling methods. The downside is that without this latent upscaling + repainting approach, I lose some of the fine details I’m aiming for.

Features

• 3+ outfits

Issues

• There’s interference between clothing concepts — the default outfit is overfitted, causing weaker performance on the other outfits.

• Control over clothing details is also not stable enough.

• Some details, such as Tacet Mark, haven’t been learned properly.

• When the face becomes blurry, consider using ADetailer to fix it

TRIGGER WORDS

Default

j1nhs1, silver hair, long hair, blunt bangs, twintails, light blue hairbow, gold hairclip, silver eyes, black underdress, white overcoat, cyan cape, armband, wristband, white thighboots, 

Under Dress

j1nhs1, silver hair, long hair, blunt bangs, twintails, light blue hairbow, gold hairclip, silver eyes, black underdress, barefoot,

Hanfu Dress

j1nhs1, silver hair, long hair, blunt bangs, twintails, light blue hairbow, gold hairclip, silver eyes, blue chest-skirt, sheer-robe, orange waist ribbon, orange sash around arms, 

Requirements

Checkpoints: Anima Preview 2

Recommended Tools: ComfyUI, Forge Neo

Recommended Step: 20~30

Recommended weight: 0.6~1

Recommended size: 768*1344, 832*1216, 896*1156, 1024*1024, 1344*768, 1216*8

32, 1156*896

I’ve set the gallery moderation of this post to “R”, so explicit content may not appear on this page.