StableDiffusionWebUI: Linux Installation

Preface

The previous section covered how to install Stable Diffusion on Mac: StableDiffusionWebUI: Mac Installation

The Linux installation is quite similar.

This section introduces how to install SD on Linux.

Server

Since the installation process requires many dependencies,

it is recommended to use a Hong Kong spot GPU server for the installation.

After installation, create a server image and sync it to other regions.

Make sure to check the option to automatically install GPU drivers when configuring the server. This avoids many subsequent issues.

This way, the GPU drivers will be automatically installed when you first enter the server.

If the drivers are not installed automatically, you can install them manually:

https://cloud.tencent.com/document/product/560/8048?from_cn_redirect=1

Installation

Update Server Software

It is best to update first to avoid unexpected issues.

sudo apt update
sudo apt upgrade

Install Dependencies

sudo apt install wget git python3 python3-venv libgl1 libglib2.0-0

Install tcmalloc

During the subsequent installation, you will be prompted whether to install tcmalloc to improve memory allocation. You can install it in advance.

sudo apt install google-perftools

Download Installation Script

mkdir sdwebui && cd sdwebui
wget -q https://raw.githubusercontent.com/AUTOMATIC1111/stable-diffusion-webui/master/webui.sh
sudo chmod +x webui.sh

Run the Script

# For the first launch, run without additional parameters to let sdwebui install dependencies
./webui.sh

Startup Complete

After a long wait, sdwebui starts successfully. Subsequent startups will be much faster.

SDWebUI

|—Link: https://sdwebui.ai/?from=vq

|—Authentic Stable Diffusion WebUI V1.10.1~

|—Top-tier 4090 GPU, premium members get exclusive single 4090 card for image generation~

|—Supports Mac client, Windows client, and Web online usage~

|—Innovative instant model transfer feature, instantly use local models~

|—Supports txt2img, img2img, extras, and PNG info

|—Supports CLIP interrogation and DeepBooru interrogation

|—Supports Hires.fix with 21 built-in upscaling models

|—Supports ADetailer with 11 built-in repair models

|—Supports ControlNet with 44 built-in preprocessors and 68 built-in models

|—Supports super-resolution upscaling (Tiled Diffusion, Demo Fusion, Tiled VAE)

|—Supports old photo restoration with 21 built-in upscaling models, plus GFPGAN and CodeFormer restoration models

|—Built-in 2 common Styles, 11 popular Checkpoint models, 7 popular LoRA models, 14 common embeddings

|—Built-in oldsix prompt plugin

|—Built-in 23-lesson beginner tutorial