Flux-Midjourney-Studio-LoRA开源模型 - 按文本提示生成特定风格图像

首页

Flux Midjourney Studio LoRA

由 strangerzonehf 开发

Flux-Midjourney-Studio-LoRA 是一个基于 LoRA 技术的文本到图像生成模型，能够根据文本提示生成特定风格的图像。

图像生成开源协议:其他 #Midjourney风格微调 #人像摄影生成 #LoRA图像控制

下载量 104

发布时间 : 5/3/2025

模型简介

该模型通过微调 black-forest-labs/FLUX.1-dev 基础模型，利用 LoRA 技术生成符合特定文本提示的图像，适用于图像创作领域。

模型特点

LoRA 微调技术

通过 LoRA 技术对基础模型进行微调，能够生成符合特定文本提示的图像。

多样化的图像生成参数

提供多种图像生成参数设置，可根据需求调整图像效果。

高质量图像生成

能够生成具有特定风格和细节的高质量图像，适用于专业图像创作。

模型能力

文本到图像生成

特定风格图像生成

高质量图像渲染

使用案例

图像创作

人物肖像生成

根据文本提示生成具有特定外貌特征和风格的人物肖像。

生成的人物肖像具有高度细节和艺术风格。

场景生成

根据文本提示生成特定场景的图像，如风景、室内设计等。

生成的场景图像具有丰富的细节和逼真的效果。

🚀 Flux-Midjourney-Studio-LoRA

Flux-Midjourney-Studio-LoRA 是一个文本到图像的 LoRA 模型，借助它，你可以根据文本提示生成特定风格的图像，在图像创作领域有着广泛的应用价值。

示例展示

文本提示	输出图像
mj x1, a close-up shot of a man with short platinum blonde hair and olive skin, dressed in a black crew neck t-shirt. He has a silver nose ring and a small teardrop tattoo under his left eye. The studio lighting casts a soft shadow under his chin, giving dimension to his jawline. The background is a gradient from pale blue to white.	查看
mj x1, A close-up of a young man with dark brown eyes and wavy black hair. He is wearing a dark green trench coat with a high collar and a light brown scarf around his neck. The backdrop is a cloudy gray, adding an air of mystery to the scene.	查看
mj x1, a close-up shot of a young mans face is adorned with a beige baseball cap adorned with red lettering. The mans eyes are a piercing blue, and he is wearing a pink t-shirt. His hair is dark brown, adding a touch of texture to his face. The backdrop is a vibrant shade of blue, creating a stark contrast to the mans head and the cap.	查看
mj x1, cinematic shot, a medium-sized man stands in the foreground of the frame. He is dressed in a black jacket, black pants, and a black backpack. His hair is short and dark, and hes looking off to the side. The backdrop is a stark white wall, and the ceiling is adorned with metal pipes and beams. To the left of the man, a man with dark hair and glasses is looking off into the distance.	查看

Untitled design.png

✨ 主要特性

基于 black-forest-labs/FLUX.1-dev 基础模型，通过 LoRA 技术进行微调，能生成符合特定文本提示的图像。
提供多种图像生成参数设置，可根据需求调整图像效果。

📚 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	30 & 3900
Epoch	25	Save Every N Epochs	1 [90]

标注：采用 florence2 - en（自然语言 & 英语）进行标注。
训练使用的总图像数：27 张 [Midjourney 生成的合成图像]

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

📦 安装指南

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-Midjourney-Studio-LoRA"
trigger_word = "mj x1"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

💻 使用示例

基础用法

在生成图像时，你需要使用 mj x1 作为触发词。以下是一个简单的使用示例：

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-Midjourney-Studio-LoRA"
trigger_word = "mj x1"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

prompt = f"{trigger_word}, a beautiful landscape"
image = pipe(prompt).images[0]
image.save("output.png")