Star Guardian Poppy

Version 1.0 - Initial Release

This is the first public version of the Star Guardian Poppy Style/Outfit LoRA.

Training Base: The model was trained on a focused dataset of 29 images using NoobAI XL Vpred 1.0 as the base model.
Training Type: This is a UNet-Only LoRA. It is designed to replicate the visuals (outfit, hair, hammer) of the character and relies on your prompt to create the underlying anatomy and features.
Key Characteristics:
- The LoRA was trained with a 2:1 alpha-to-dimension ratio (α=64, dim=32), making it very potent. It is crucial to use lower weights (e.g., 0.6-0.8) to avoid oversaturation.
- Advanced training techniques, including a very high min_snr_gamma of 80, were used to maximize detail and adherence to the source style.

The goal of this initial version is to provide a strong, high-quality LoRA for generating the full Star Guardian Poppy outfit with excellent accuracy.

Description:

This LoRA is dedicated to recreating the vibrant and heroic Star Guardian skin for Poppy from League of Legends.

This is a UNet-only LoRA (network_train_unet_only = true), meaning it was trained exclusively on the visual aspects of the character's design. It does not teach the model a new "concept" of Poppy but instead excels at applying her Star Guardian outfit, hairstyle, and signature hammer over a well-prompted character.

For optimal compatibility, this LoRA was trained on NoobAI XL Vpred 1.0 and works best with v-prediction models.

Usage Instructions

Activation & Prompting

Since the text encoder was not trained, you must describe the character in your prompt. Use the primary trigger word, then add descriptive tags for her appearance and outfit.

Primary Trigger: sgpoppy
Essential Character Tags: yordle, blue skin, purple hair, twintails, blue eyes, pointy ears, fang
Essential Outfit Tags: star guardian uniform, armor, breastplate, skirt, blue gloves, pantyhose, white legwear, kneeboots
Weapon Tag: hammer

Important: LoRA Weight

This LoRA was trained with an alpha (64) that is double its dimension (32). This makes the LoRA very strong.

Works plenty fine with a weight of 1.0.
- Example: <lora:SGPoppy:1.0>

Recommended Settings

Model: Any SDXL v-prediction model (e.g., NoobAI-XL, Illustrious-XL).
Sampler: Euler a CFG++
Steps: 20
CFG Scale: 1.4-1.7

Outfit & Style Description

This LoRA accurately reproduces the key elements of Poppy's Star Guardian design:

The Outfit: Captures her distinct blue, white, and gold magical girl uniform, complete with the star emblem, large bows, and armored gauntlets/boots.
Character Features: Reliably generates her yordle physique, vibrant blue skin, and signature purple hair styled in large twintails.
The Starlight Hammer: Excellently reproduces her iconic, oversized golden hammer, maintaining its grand scale in proportion to her small stature.

The style is sharp and illustrative, perfect for creating high-impact character art that reflects Poppy's determined and heroic personality.

Training Details

This LoRA was trained on a focused dataset of 29 images with the following notable parameters:

Base Model: NoobAI XL Vpred 1.0
Resolution: 1024x1024 (with bucketing enabled)
Network Rank (Dimension): 32
Network Alpha: 64
- Note: A 2:1 alpha-to-dim ratio was used, amplifying the LoRA's trained effect. This is why lower weights are recommended during inference.
Training Target: UNet Only
Optimizer: AdamW8bit
Learning Rate: 4e-4 (4×10−4)
LR Scheduler: cosine_with_restarts (5 cycles)
Advanced Noise Scheduling: An extremely high min_snr_gamma = 80 was used alongside zero_terminal_snr = true, forcing the model to focus intensely on high-level composition and detail.
Max Training Epochs: 50

Model Type	LORA
Base Model	NoobAI
Published	9/9/2025
Trained Words	sgpoppy yordle, blue skin, purple hair, twintails, blue eyes, pointy ears, fang star guardian uniform, armor, breastplate, skirt, blue gloves, pantyhose, white legwear, kneeboots hammer

Details

Download Files

About this version

Model description