Anime Detailer Xl Lora

🚀 Anime Detailer XL LoRA

Anime Detailer XL LoRA是一款先進的LoRA適配器，可與Animagine XL 2.0配合使用。它專注於概念調製，能讓用戶調整生成的動漫風格圖像的細節程度，通過操作概念滑塊，用戶可以創建從高度細節到更抽象表現的圖像。

🚀 快速開始

安裝依賴

確保安裝最新的 diffusers 庫以及其他必要的包：

pip install diffusers --upgrade
pip install transformers accelerate safetensors

運行示例代碼

以下Python腳本展示瞭如何將LoRA與Animagine XL 2.0結合使用。默認調度器為 EulerAncestralDiscreteScheduler，為清晰起見，也可顯式定義。

import torch
from diffusers import (
    StableDiffusionXLPipeline, 
    EulerAncestralDiscreteScheduler,
    AutoencoderKL
)

# Initialize LoRA model and weights
lora_model_id = "Linaqruf/anime-detailer-xl-lora"
lora_filename = "anime-detailer-xl.safetensors"
lora_scale_slider = 2 # -2 for less detailed result

# Load VAE component
vae = AutoencoderKL.from_pretrained(
    "madebyollin/sdxl-vae-fp16-fix", 
    torch_dtype=torch.float16
)

# Configure the pipeline
pipe = StableDiffusionXLPipeline.from_pretrained(
    "Linaqruf/animagine-xl-2.0", 
    vae=vae,
    torch_dtype=torch.float16, 
    use_safetensors=True, 
    variant="fp16"
)
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.to('cuda')

# Load and fuse LoRA weights
pipe.load_lora_weights(lora_model_id, weight_name=lora_filename)
pipe.fuse_lora(lora_scale=lora_scale_slider)

# Define prompts and generate image
prompt = "face focus, cute, masterpiece, best quality, 1girl, green hair, sweater, looking at viewer, upper body, beanie, outdoors, night, turtleneck"
negative_prompt = "lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry"

image = pipe(
    prompt, 
    negative_prompt=negative_prompt, 
    width=1024,
    height=1024,
    guidance_scale=12,
    num_inference_steps=50
).images[0]

# Unfuse LoRA before saving the image
pipe.unfuse_lora()
image.save("anime_girl.png")

✨ 主要特性

Anime Detailer XL LoRA是一款前沿的LoRA適配器，專為與Animagine XL 2.0協同工作而設計。該獨特模型專注於概念調製，使用戶能夠調整生成的動漫風格圖像的細節程度。通過操作概念滑塊，用戶可以創建從高度詳細到更抽象表現的圖像。

📦 安裝指南

確保安裝最新的 diffusers 庫以及其他必要的包：

pip install diffusers --upgrade
pip install transformers accelerate safetensors

💻 使用示例

基礎用法

import torch
from diffusers import (
    StableDiffusionXLPipeline, 
    EulerAncestralDiscreteScheduler,
    AutoencoderKL
)

# Initialize LoRA model and weights
lora_model_id = "Linaqruf/anime-detailer-xl-lora"
lora_filename = "anime-detailer-xl.safetensors"
lora_scale_slider = 2 # -2 for less detailed result

# Load VAE component
vae = AutoencoderKL.from_pretrained(
    "madebyollin/sdxl-vae-fp16-fix", 
    torch_dtype=torch.float16
)

# Configure the pipeline
pipe = StableDiffusionXLPipeline.from_pretrained(
    "Linaqruf/animagine-xl-2.0", 
    vae=vae,
    torch_dtype=torch.float16, 
    use_safetensors=True, 
    variant="fp16"
)
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.to('cuda')

# Load and fuse LoRA weights
pipe.load_lora_weights(lora_model_id, weight_name=lora_filename)
pipe.fuse_lora(lora_scale=lora_scale_slider)

# Define prompts and generate image
prompt = "face focus, cute, masterpiece, best quality, 1girl, green hair, sweater, looking at viewer, upper body, beanie, outdoors, night, turtleneck"
negative_prompt = "lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry"

image = pipe(
    prompt, 
    negative_prompt=negative_prompt, 
    width=1024,
    height=1024,
    guidance_scale=12,
    num_inference_steps=50
).images[0]

# Unfuse LoRA before saving the image
pipe.unfuse_lora()
image.save("anime_girl.png")

📚 詳細文檔

模型詳情

🔧 技術細節

Anime Detailer XL LoRA作為一個LoRA適配器，與Animagine XL 2.0結合使用，通過概念調製來控制生成的動漫圖像的細節程度。它利用了 diffusers 庫中的 StableDiffusionXLPipeline 進行圖像生成，同時使用 EulerAncestralDiscreteScheduler 作為調度器。在代碼實現中，首先加載VAE組件，然後配置管道，接著加載並融合LoRA權重，最後根據定義的提示生成圖像。在保存圖像之前，需要解除LoRA的融合。

📄 許可證

本項目採用 CreativeML Open RAIL++-M License 許可證。

致謝

我們的項目受益於以下重要工作：

• Erasing Concepts from Diffusion Models 作者：Rohit Gandikota等人
• LECO 作者：p1atdev
• AI Toolkit 作者：Ostris