case2

最新推荐文章于 2024-09-13 14:33:56 发布

kfrealme

最新推荐文章于 2024-09-13 14:33:56 发布

阅读量1.5k

点赞数 44

文章标签：人工智能

本文链接：https://blog.csdn.net/kfrealme/article/details/138869127

版权

ComfyUI Example 2

琉璃效果复现

建议直接看下方安装失败

详情：

IMAGE TO 💎JADE STYLE | ComfyUI Workflow

准备

checkpoint 模型：

Juggernaut XL - Jugg_X_RunDiffusion_Hyper | Stable Diffusion Checkpoint | Civitai

lora模型：

glazed girl

https://civitai.com/models/140681/xlglazed-girl-jade-glass-ceramic-and-other-textures

sdxl_glass

https://civitai.com/models/11203/glass-sculptures

jade

https://civitai.com/models/148809/sdxl-or-jadeite-cabbage-or-chinese-treasure

放大模型

4x-Ultrasharp - 4x-UltraSharp v1.0 | Stable Diffusion Upscaler | Civitai

开始

打开工作流

- 安装缺失的节点

GitHub - Fannovel16/comfyui_controlnet_aux: ComfyUI’s ControlNet Auxiliary Preprocessors

cd  custom_nodes 
git clone https://github.com/Fannovel16/comfyui_controlnet_aux/
cd comfyui_controlnet_aux
pip install -r requirements.txt

GitHub - WASasquatch/was-node-suite-comfyui: An extensive node suite for ComfyUI with over 190 new nodes

cd  custom_nodes 
git clone https://github.com/WASasquatch/was-node-suite-comfyui/
cd was-node-suite-comfyui
pip install -r requirements.txt

GitHub - EllangoK/ComfyUI-post-processing-nodes: A collection of Post Processing Nodes for ComfyUI, which enable a variety of cool image effects

git clone https://github.com/EllangoK/ComfyUI-post-processing-nodes/

GitHub - Suzie1/ComfyUI_Comfyroll_CustomNodes: Custom nodes for SDXL and SD1.5 including Multi-ControlNet, LoRA, Aspect Ratio, Process Switches, and many more nodes.

git clone https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes.git

GitHub - cubiq/ComfyUI_IPAdapter_plus

git clone https://github.com/cubiq/ComfyUI_IPAdapter_plus.git

Page not found · GitHub · GitHub

git clone https://github.com/sipherxyz/comfyui-art-venture.git

GitHub - Kosinkadink/ComfyUI-Advanced-ControlNet: ControlNet scheduling and masking nodes with sliding context support

git clone https://github.com/Kosinkadink/ComfyUI-Advanced-ControlNet.git

GitHub - rgthree/rgthree-comfy: Making ComfyUI more comfortable!

git clone https://github.com/rgthree/rgthree-comfy.git

GitHub - shadowcz007/comfyui-mixlab-nodes: Workflow-to-APP、ScreenShare&FloatingVideo、GPT & 3D、SpeechRecognition&TTS

git clone  https://github.com/shadowcz007/comfyui-mixlab-nodes.git

controlnet 模型

sai_xl_canny_256lora.safetensors · lllyasviel/sd_control_collection at main

sai_xl_depth_256lora.safetensors · lllyasviel/sd_control_collection at main

clip_vision 模型

/ComfyUI/models/clip_vision
- CLIP-ViT-H-14-laion2B-s32B-b79K.safetensors, download and rename
- CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors, download and rename

运行过程中还要下载模型，

没错安装失败

本地安装失败。白下几十个 G,
切换在线使用复现琉璃效果

网站：eSheep(内测中) - 一站式的AIGC社区

工作流下载地址： IMAGE TO 💎JADE STYLE | ComfyUI Workflow

在线启动：

更换模型，checkpoint, lora, 放大模型，
上传图片，两个地方需要上传，
开始，
成功出图
对比：图片地址：eSheep(内测中) - 一站式的AIGC社区

只要成功复现就说明步骤没问题，工作流也没问题。

同样的参数，在我的设备不是不能完全复现就是直接安装失败，模型下载了几十个 g, 靠北熬

好，好歹成功复现，增强学习信心。

后面准备一个案例学习一个 cmfyui功能