🚀 Flux-Chill-Guy-Zone [他无所不能]
Flux-Chill-Guy-Zone模型能够基于输入的文本描述生成各种风格的图像,尤其擅长生成“Chill Guy”这一卡通角色在不同场景下的图像,为图像创作提供了丰富的可能性。
🚀 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 = "prithivMLmods/Flux-Chill-Guy-Zone"
trigger_word = "chill guy"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
触发词使用
你应该使用 chill guy
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,可在“Files & versions”标签中获取。
✨ 主要特性
- 多样化场景生成:可以生成“Chill Guy”在城市、浴室、公园、海边等多种不同场景下的图像。
- 丰富细节表现:能够细致描绘角色的穿着、姿态、表情以及场景中的背景元素。
📦 安装指南
暂未提供具体安装步骤,可参考上述“快速开始”中的环境设置部分。
💻 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 = "prithivMLmods/Flux-Chill-Guy-Zone"
trigger_word = "chill guy"
pipe.load_lora_weights(lora_repo)
device = torch.device("cuda")
pipe.to(device)
prompt = "chill guy, a Peanuts character is standing on a rooftop, facing the viewer. He is wearing a white shirt, white pants, and red and white shoes. His ears are perked up, and his eyes are squinted. His mouth is slightly open, as if he is smiling. The backdrop of the scene is a cityscape, with skyscrapers and a hot air balloon in the foreground. The sky is a deep blue, with a few wispy clouds, adding a touch of color to the scene."
image = pipe(prompt).images[0]
image.save("output.png")
📚 详细文档
模型描述
prithivMLmods/Flux-Chill-Guy-Zone
图像参数处理如下:
参数 |
详情 |
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 |
17 & 2600 |
Epoch |
15 |
Save Every N Epochs |
1 |
标注:florence2 - en(自然语言 & 英语)
训练使用的总图像数:18
最佳尺寸
- 768 x 1024(最佳)
- 1024 x 1024(默认)
📄 许可证
本模型使用的许可证为 creativeml-openrail-m
。
⚠️ 重要提示
该模型仍处于训练阶段,这不是最终版本,可能会包含瑕疵,并且在某些情况下表现不佳。