ComfyUI+krita AI实时绘画安装部署方法

实时AI绘画的一键整合包实时AI绘画工作流，使用Krita+ComfyUI。笔记通过本人验证。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

概要：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• 整合了Krita+ComfyUI
• 提供了相关压缩包下载链接
• 绘画简要教超
• 官方ComfyUI介绍参考：ComfyUI介绍(官方直译)详细部署教程和使用

下载Krita+ComfyUI使用教程：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

下载压缩包 下载地址：链接：https://pan.baidu.com/s/1vCgioW1ObIYmesDFurSzyA?pwd=6666文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

krita绘画软件官网地址https://krita.org/en/download/krita-desktop/文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

krita-ai-diffusion 插件：https://github.com/Acly/krita-ai-diffusion文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• 解压，最好不要有中文路径，解压后，文件如下图所示

1. 第一个压缩包为ComfyUI整合包，再解压此文件夹

修改配置，实现与Stable Diffusion 中的模型库文件共享（避免重复下载资源）：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

stable-diffusion-webui 如何跟ComfyUI之间共享模型，以下说明配置方法：将extra_model_paths.yaml.example文件复制一份，然后通过文本编辑器打开副本文件。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

其中ComfyUI-master中，修改extra_model_paths.yaml:文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

该文件中的更改和保存项为（最好将原本Windows路径里的\ ，改为/ ）：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

a111:文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

base_path: F:\qiuye_V4.5\Stable Diffusion #path/to/stable-diffusion-webui/文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

checkpoints: models/Stable-diffusion configs: models/Stable-diffusion vae: models/VAE loras: | models/Lora models/LyCORIS upscale_models: | models/ESRGAN models/RealESRGAN models/SwinIR embeddings: embeddings hypernetworks: models/hypernetworks controlnet:extensions\sd-webui-controlnet\models #models/ControlNet文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

安装其中的ComfyUI-Manager插件工具：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

ComfyUI-Manager插件的工具下载地址：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

https://github.com/ltdrdata/ComfyUI-Manager文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

1）下载解压文件，复制粘贴到custom_nodes中。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

2）方法2：提前装好git工具；用git命令下载到对应环境下。（在custom_nodes目录下，输入cmd，然后输入 git clone http：//地址。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

启动ComfyUI：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

1. 双击run_nvidia _gpu.bat启动ComfyUI

git 命令安装前往*\ComfyUI\custom_nodes目录在文件路径处输入cmd 按下回车会打开cmd命令行。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

通过该命令安装：git clone https://github.com/ltdrdata/ComfyUI-Manager.git下图表示安装成功，重启ComfyUI。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

配置使用Stable Diffusion的模型库路径：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

2.第2个安装包，双击安装krita-x64-5.2.1-setup.exe，一直点击下一步即可。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• 桌面打开安装好的krita软件，从工具->脚本->从文件导入 Python 插件...,选择解压后的krita_ai_diffusion-1.7.1.zip文件

• 然后重启krita软件即可。
• 到目前为止，所有环境具体就准备好了。
• 点击新建文件，像素设置为500*500：

Krita连接配置出现缺模型：

提示缺少CLIPVision model pytorch_model.bin文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

pytorch_model.bin官方下载地址：https://huggingface.co/h94/IP-Adapter/tree/main/models/image_encoder文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

提示： ip-adapter_sd15.safetensors, ip-adapter_sdxl_vit-h.safetensors. Make sure to download the model and place it in the ComfyUI/custom_nodes/ComfyUI_IPAdapter_plus/models folder.文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

下载地址：ip-adapter_sd15.safetensors 链接: https://pan.baidu.com/s/1274u97ZXw64mhtsvkFLUOw?pwd=6666 提取码: 6666 https://huggingface.co/h94/IP-Adapter/tree/main/models ;文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

ip-adapter_sdxl_vit-h.safetensors https://huggingface.co/h94/IP-Adapter/tree/main/models文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

ComfyUI汉化插件

AIGODLIKE-ComfyUI-Translation 用于 ComfyUI 多语言翻译的插件，该插件实现了常驻菜单栏/搜索栏/右键上下文菜单/节点等的翻译。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

通过Manager安装AIGODLIKE-ComfyUI-Translation，没用git或者访问github下载缓慢请往下看找到手动安装。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Krita 5.2.1 下载地址：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

https://mirrors.bfsu.edu.cn/kde/stable/krita/5.2.1/krita-x64-5.2.1-setup.exe文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

IP-Adapter下载地址：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

【模型介绍】这个可以让你玩出花儿来的模型-IP-Adapter文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

视频krita和ComfyUI联动：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

https://xueshu.fun/3244/文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

当提示“connected”时，则表示，Krita与ComfyUI连接匹配成功可以实现实时动态绘画。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Krita与ComfyUI的演示示意可以查看：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

ComfyUI介绍(官方直译)详细部署教程和使用文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Krita连接ComfyUI时需要安装的自定义节点（需要访问github）：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Install custom nodes according to the instructions of the respective projects, or use ComfyUI Manager.文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• ControlNet preprocessors
• IP-Adapter
• Ultimate SD Upscale
• External tooling nodes

需要的模型文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Download models to the paths indicated below. If you are using extra_model_paths.yml, those will also work. Shared models are always required, and at least one of SD1.5 and SDXL is needed.文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

Shared文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• Clip-Vision to models/clip_vision/SD1.5
• NMKD Superscale SP_178000_G to models/upscale_models

SD 1.5文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• ControlNet inpaint to models/controlnet
• ControlNet tile to models/controlnet
• IP-Adapter (SD1.5) to custom_nodes/ComfyUI_IPAdapter_plus/models
• LCM-LoRA (SD1.5) to models/loras/lcm-lora-sdv1-5.safetensors _ ！*一定记住改名！

SD XL文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/60529.html

• IP-Adapter (SDXL) to custom_nodes/ComfyUI_IPAdapter_plus/models
• LCM-LoRA (SDXL) to models/loras/lcm-lora-sdxl.safetensors rename!
• Checkpoints The following checkpoints are used by the default styles:
• Realistic Vision
• DreamShaper
• JuggernautXL