nodes/smthemex-ComfyUI_MS_Diffusion
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
smthemex
2024-09-07 14:33:10 +08:00
committed by GitHub
parent 1bc5b0612f
commit 7c716cedbc
1 changed files with 10 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+10 -46
View File
@@ -6,7 +6,7 @@ MS-Diffusion origin From: [MS-Diffusion](https://github.com/MS-Diffusion/MS-Diff
**NEW Update**
---
* 2024/09/06:fix runway error
* 2024/09/06:fix runway error/load single controlnet now /首次加载不再需要连外网。
**Previous updates**
* del clip repo，Add comfyUI clip_vision loader/加入comfyUI的clip vision节点，不再使用 clip repo。
@@ -29,56 +29,20 @@ If the module is missing, please pip install
3 Need model
----
3.1 base model:
底模使用社区单体模型，首次使用需要下载一些必备的config文件，没连外网或者没有做镜像站映射的，肯定是用不了的。
Choose one of the three options above the model loading node, and change the rest to "none" or leave it blank. The first option is to select a standalone SDXL community model, the second option is to select a pre downloaded SDXL diffuser model, and the third option is to use the "repo_id" method
The default repo_id models are stabilityai/stable-diffusion-xl-base-1.0 (for example,you can using :G161222/RealVisXL_V4.0,sd-community/sdxl-flash...)
3.2 adapter and encoder model
必须的模型，位置如图：
Need download "ms_adapter.bin" : [link](https://huggingface.co/doge1516/MS-Diffusion/tree/main)
```
├── ComfyUI/models/
| ├──photomaker/
| ├── ms_adapter.bin
```
clip vision model （ any base from CLIP-ViT-bigG-14-laion2B-39B-b160k）
Control_img image preprocessing, please use other nodes
```
├── ComfyUI/models/
| ├──clip_vision/
| ├── clip_vision_g.safetensors or CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors
├── ComfyUI/models/checkpoints/
| ├── any SDXL weigth
├── ComfyUI/models/photomaker/
| ├── ms_adapter.bin
├── ComfyUI/models/clip_vision/
| ├── clip_vision_g.safetensors or CLIP-ViT-bigG-14-laion2B-39B-b160k.safetensors
├── ComfyUI/models/controlnet
| ├──any SDXL controlnet
```
3.3 only supports SDXL controllnet
以下是controlnet的模型和config文件，ComfyUI的用的单体controlnet模型用不了。用了太占显存了，改回pipe基本加载模式。
```
├── ComfyUI/models/diffusers/
| ├──xinsir/controlnet-openpose-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──xinsir/controlnet-scribble-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──diffusers/controlnet-canny-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──diffusers/controlnet-depth-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──/controlnet-zoe-depth-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──TheMistoAI/MistoLine
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
| ├──xinsir/controlnet-tile-sdxl-1.0
| ├── config.json
| ├── diffusion_pytorch_model.fp16.safetensors
```
Control_img image preprocessing, please use other nodes
4 using tips
---
--生成物体名称需要用[ ]括起来，有多少个物体，就要有多少张图片输入;