This LoRA must not be used to generate NSFW content.
Expand this post for the recommended Anima parameters.
Suggested tags for consistent results:
military uniform, green military uniform, long sleeves, high collar, epaulettes, belt, green pants, black boots
Recommended settings:
DPM++ 2M
Karras is prefered
Sampling Steps: 25-35
CFG Scale: 5-6
Model Weight: 0.8-1 (tested with 0.8)
Clip Skip: 2
This LoRA was trained on a small, mixed-quality dataset.
As a personal preference, I generally use ADetailer for the faces of any generation, this model gave me some issues with the face. I would recommend using a face ADetailer.
I have had success with multiple Checkpoints. Realistic checkpoints tend to age up and change body structure.
Anima Version:
tanya_kurba, blonde hair, ahoge, poytail, blue eyes, green military uniform, epaulettes, belt, green pants, black boots,If you want to add her military cap, replace ahoge and poytail with peaked cap.
I've had mixed results while using quality tags. Here are a few that I've tested and found to work across multiple checkpoints.
Generally okay to use with any checkpoint:
masterpiece, best quality, score_9, score_8, score_7, absurdres, highres, newest, Best with higher-detail checkpoints:
hyper realism, photorealistic,The realism tags improved overall quality while preserving the anime style.
The lighting was the most notable difference when using these tags.
Recommended settings:
Sampling Steps: 15-35
CFG Scale: 3.5-4
I would love to see what you create with this LoRA. If you have any feedback, I'd be be interested in hearing it.
Description
This version is trained on the same dataset as the Illustrious version, but using Anima.





