This LoRA model generates digital drawings in a vibrant anime style, featuring characters with distinct traits such as large, expressive eyes and stylish outfits. It captures dynamic poses and accessories, exemplified by a young girl holding a black assault rifle, set against a clean, minimalistic background.

Trigger word: miladyy

Sample Outputs


Sample 1	Sample 2
miladyy, This is a digital drawing in an anime style. The su...	miladyy, This is a digital drawing in a chibi style featurin...
Sample 3	Sample 4
miladyy, This is a digitally altered photograph featuring a ...	miladyy, This is a digitally created promotional image featu...

Usage

ComfyUI

Download the .safetensors file from the Files tab
Place in ComfyUI/models/loras/
Use the Load LoRA node with strength 0.8-1.0
Include miladyy in your prompt

Diffusers

python

import torch
from diffusers import FluxPipeline

pipe = FluxPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-dev",
    torch_dtype=torch.bfloat16
)
pipe.load_lora_weights("ms2stationthis/miladyy")
pipe.to("cuda")

image = pipe(
    prompt="miladyy portrait, soft lighting, detailed",
    guidance_scale=3.5-4.0,
    num_inference_steps=20-30,
    generator=torch.Generator("cuda").manual_seed(42)
).images[0]

image.save("output.png")

Recommended Settings

Parameter	Value
LoRA Strength	0.8-1.0
Guidance Scale	3.5-4.0
Inference Steps	20-30
Resolution	1024x1024

Example Prompts

miladyy blonde-haired girl with large
miladyy expressive blue eyes
miladyy wearing a white visor with a red "c" logo
miladyy a yellow blouse

Training Details

Base model: black-forest-labs/FLUX.1-dev
Training steps: 4000
Trigger word: miladyy
Network: LoRA rank 32, alpha 32
Optimizer: adamw8bit, lr 1e-4
Precision: float16
Resolution: 512, 768, 1024 (multi-res bucketed)

About

Trained using StationThis — an AI creative platform powered by $MS2. Train your own LoRAs via @stationthisbot on Telegram.

_{Generated by StationThis Training Pipeline}

noema-art/miladyy

miladyy