Flux Ghibli Art LoRA
F
Flux Ghibli Art LoRA
由 strangerzonehf 开发
基于FLUX.1-dev基础模型的LoRA微调模型,专门用于生成吉卜力风格的艺术图像。
下载量 3,802
发布时间 : 12/6/2024
模型简介
该模型通过LoRA技术对基础扩散模型进行微调,能够根据文本描述生成具有吉卜力工作室标志性风格的图像作品。
模型特点
吉卜力风格艺术生成
专门优化用于生成具有吉卜力工作室标志性风格的图像作品
LoRA微调技术
采用LoRA(Low-Rank Adaptation)技术对基础模型进行高效微调
多场景适应
能够处理各种场景描述,包括自然景观、人物活动等吉卜力风格常见主题
模型能力
文本到图像生成
艺术风格转换
场景可视化
使用案例
艺术创作
概念艺术生成
为动画或游戏项目快速生成吉卜力风格的概念艺术
可生成符合吉卜力美学标准的场景和角色设计
插画创作辅助
帮助插画师快速构思和实现吉卜力风格的画面
提供风格一致的图像参考和灵感
娱乐应用
个性化艺术创作
让用户通过简单文本描述创建属于自己的吉卜力风格图像
用户可自定义场景、角色等元素生成独特艺术作品
🚀 Flux-Ghibli-Art-LoRA
Flux-Ghibli-Art-LoRA是一个基于LoRA技术的图像生成模型,它以特定风格(吉卜力艺术风格)为特色,能根据文本描述生成具有该风格的图像,为图像创作带来更多可能性。
🚀 快速开始
环境设置
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "strangerzonehf/Flux-Ghibli-Art-LoRA"
trigger_word = "Ghibli Art"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
触发词使用
你应该使用 Ghibli Art
来触发图像生成。
模型下载
此模型的权重以Safetensors格式提供。
点击下载(在“文件与版本”选项卡中)。
✨ 主要特性
- 风格独特:能够生成具有吉卜力艺术风格的图像,带来独特的视觉体验。
- 参数可配:提供了丰富的图像处理参数,可根据需求进行调整。
📦 安装指南
暂未提供具体安装步骤,可参考上述快速开始部分的环境设置代码。
💻 使用示例
基础用法
import torch
from pipelines import DiffusionPipeline
base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)
lora_repo = "strangerzonehf/Flux-Ghibli-Art-LoRA"
trigger_word = "Ghibli Art"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
# 以下可根据需求添加生成图像的代码
📚 详细文档
模型描述
图像处理参数
参数 | 值 | 参数 | 值 |
---|---|---|---|
LR Scheduler | constant | Noise Offset | 0.03 |
Optimizer | AdamW | Multires Noise Discount | 0.1 |
Network Dim | 64 | Multires Noise Iterations | 10 |
Network Alpha | 32 | Repeat & Steps | 15 & 2400 |
Epoch | 15 | Save Every N Epochs | 1 |
标注方式:florence2 - en(自然语言 & 英语)
用于训练的图像总数:20
最佳尺寸与推理
尺寸 | 宽高比 | 推荐情况 |
---|---|---|
1280 x 832 | 3:2 | 最佳 |
1024 x 1024 | 1:1 | 默认 |
推理范围
- 推荐推理步数:30–35
🔧 技术细节
该模型基于 black-forest-labs/FLUX.1-dev
基础模型,使用LoRA技术进行微调。通过设置特定的图像处理参数,如学习率调度器、优化器等,结合标注好的图像数据进行训练,从而实现生成具有吉卜力艺术风格图像的功能。
📄 许可证
本模型使用 openrail
许可证。
示例展示
以下是一些使用该模型生成的图像示例:
- 输入文本:'Ghibli Art, a vibrant orange van is parked on a rocky hillside. The van is adorned with a white roof, adorned with an orange patch, adding a pop of color to the scene. A man in a yellow hooded sweatshirt and jeans is standing on the roof of the van, looking out at the view of the mountain range. The sky is a deep blue, dotted with fluffy white clouds. To the right of the man, a large brown backpack is draped over the vans head. The mountain range is covered in snow, adding to the authenticity of the image.' 输出图像:点击查看
- 输入文本:'Ghibli Art, a cartoon-like scene features a young man with short brown hair, dressed in a white t-shirt, brown pants, and brown boots, standing on a cliff overlooking a valley. He is holding a map in his right hand, and a brown backpack on his left shoulder. The valley is filled with lush green trees, and the valley is bordered by a mountain range. The sky is a deep blue, dotted with white clouds, adding a pop of color to the scene.' 输出图像:点击查看
- 输入文本:'Ghibli Art, a young boy is riding a bicycle on a wet, rainy day. The boy is dressed in a bright orange shirt, gray pants, and a red backpack. His helmet is black, and he is facing the right side of the frame. The bicycle is casting a shadow on the water, and the water is reflecting the boys reflection. To the left of the boy, a field of yellow corn is growing, and to the right of the field is a house with a red roof. The sky is filled with dark gray clouds, adding a touch of color to the scene.' 输出图像:点击查看
- 输入文本:'Ghibli Art, Captured at eye-level on a vibrant day, a man in a blue shirt, and a straw hat is standing in a field of wheat. The man is holding a rifle in his right hand, while his left hand is resting on a wooden fence. In the distance, a brown barn with a brown roof is seen, nestled in the field. The barn is surrounded by trees and a metal tower. The sky is a deep blue, dotted with white clouds, adding a touch of color to the scene.' 输出图像:点击查看
- 输入文本:'Ghibli Art: A bustling village square on a sunny afternoon. A boy in a red scarf and green jacket rides a vintage bicycle through the cobblestone streets, carrying a basket full of bread loaves. To his left, a lively market is set up with colorful stalls, while to his right, a fountain with a bronze statue glistens in the sunlight. The sky is a vivid blue with cotton-like clouds, and the buildings are adorned with flowering vines and wooden shutters.' 输出图像:点击查看
- 输入文本:'Ghibli Art: A young girl with short black hair and a yellow raincoat stands under a red umbrella in the middle of a cobblestone street. Rain falls steadily, creating ripples in puddles around her feet. Behind her, a small bakery with a warm, glowing light spills onto the street, and its wooden sign sways gently in the wind. The sky is heavy with dark gray clouds, and a single beam of light breaks through, illuminating the scene.' 输出图像:点击查看
- 输入文本:'Ghibli Art: A narrow mountain path shrouded in rain and mist. A traveler in a wide-brimmed straw hat and a long green cloak leans on a wooden staff as they walk. Water cascades down the rocky slopes, forming small streams that crisscross the path. The dark clouds above roll ominously, and a single bird soars against the turbulent sky, adding a sense of movement to the stillness.' 输出图像:点击查看
- 输入文本:'Ghibli Art: A vibrant spring meadow at dawn. A young woman in a lavender dress and straw hat is painting on an easel under a large cherry blossom tree. Pink petals gently fall around her, and a curious white rabbit sits nearby, watching. The meadow is dotted with yellow and purple wildflowers, and a small brook winds its way through the scene. In the background, a quaint cottage with a stone chimney is nestled among rolling green hills' 输出图像:点击查看
- 输入文本:'Ghibli Art: A young boy with tousled black hair sits on a grassy hill under a dark, star-filled sky. He wears a light blue sweater and brown pants, with a small telescope set up in front of him. Around him, fireflies dance in the cool night air, adding a magical glow. A crescent moon hangs low on the horizon, and in the distance, the faint silhouette of a village is visible against the hills.' 输出图像:点击查看
Stable Diffusion V1 5
Openrail
稳定扩散是一种潜在的文本到图像扩散模型,能够根据任何文本输入生成逼真的图像。
图像生成
S
stable-diffusion-v1-5
3.7M
518
Stable Diffusion Inpainting
Openrail
基于稳定扩散的文本到图像生成模型,具备图像修复能力
图像生成
S
stable-diffusion-v1-5
3.3M
56
Stable Diffusion Xl Base 1.0
SDXL 1.0是基于扩散的文本生成图像模型,采用专家集成的潜在扩散流程,支持高分辨率图像生成
图像生成
S
stabilityai
2.4M
6,545
Stable Diffusion V1 4
Openrail
稳定扩散是一种潜在文本到图像扩散模型,能够根据任意文本输入生成逼真图像。
图像生成
S
CompVis
1.7M
6,778
Stable Diffusion Xl Refiner 1.0
SD-XL 1.0优化器模型是Stability AI开发的图像生成模型,专为提升SDXL基础模型生成的图像质量而设计,特别擅长最终去噪步骤处理。
图像生成
S
stabilityai
1.1M
1,882
Stable Diffusion 2 1
基于扩散的文本生成图像模型,支持通过文本提示生成和修改图像
图像生成
S
stabilityai
948.75k
3,966
Stable Diffusion Xl 1.0 Inpainting 0.1
基于Stable Diffusion XL的潜在文本到图像扩散模型,具备通过遮罩进行图像修复的功能
图像生成
S
diffusers
673.14k
334
Stable Diffusion 2 Base
基于扩散的文生图模型,可根据文本提示生成高质量图像
图像生成
S
stabilityai
613.60k
349
Playground V2.5 1024px Aesthetic
其他
开源文生图模型,能生成1024x1024分辨率及多种纵横比的美学图像,在美学质量上处于开源领域领先地位。
图像生成
P
playgroundai
554.94k
723
Sd Turbo
SD-Turbo是一款高速文本生成图像模型,仅需单次网络推理即可根据文本提示生成逼真图像。该模型作为研究原型发布,旨在探索小型蒸馏文本生成图像模型。
图像生成
S
stabilityai
502.82k
380
精选推荐AI模型
Llama 3 Typhoon V1.5x 8b Instruct
专为泰语设计的80亿参数指令模型,性能媲美GPT-3.5-turbo,优化了应用场景、检索增强生成、受限生成和推理任务
大型语言模型
Transformers 支持多种语言

L
scb10x
3,269
16
Cadet Tiny
Openrail
Cadet-Tiny是一个基于SODA数据集训练的超小型对话模型,专为边缘设备推理设计,体积仅为Cosmo-3B模型的2%左右。
对话系统
Transformers 英语

C
ToddGoldfarb
2,691
6
Roberta Base Chinese Extractive Qa
基于RoBERTa架构的中文抽取式问答模型,适用于从给定文本中提取答案的任务。
问答系统 中文
R
uer
2,694
98