Skip to content

Install and Run on NVidia GPUs

Jump to bottom
ClashSAN edited this page Sep 15, 2023 · 39 revisions

Automatic Installation

Windows (method 1)

A very basic guide to get Stable Diffusion web UI up and running on Windows 10/11 NVIDIA GPU.

  1. Download the sd.webui.zip from here, this package is from v1.0.0-pre we will update it to the latest webui version in step 3.
  2. Extract the zip file at your desired location.
  3. Double click the update.bat to update web UI to the latest version, wait till finish then close the window.
  4. Double click the run.bat to launch web UI, during the first launch it will download large amounts of files. After everything has been downloaded and installed correctly, you should see a message "Running on local URL: http://127.0.0.1:7860", opening the link will present you with the web UI interface.

you should be able to start generating images

Extra configurations via COMMANDLINE_ARGS

There are some configuration options that you may want apply to web UI, in order to configure these options you need to edit the launch script found at sd.webui\webui\webui-user.bat, edit the file add the selected arguments after set COMMANDLINE_ARGS= like so :

set COMMANDLINE_ARGS=--autolaunch --update-check

Each individual argument need to separated by a space, the above example well configure web UI to auto launch the browser page after it completes loading, and also check for new version of web UI at launch.

Troubleshooting

The default configuration of web UI should run on most modern GPU, but in some cases you may need some extra arguments to allow it to work properly.

  1. For GPU with less amounts of VRAM, you may need --medvram or --lowvram, these optimizations reduces VRAM requirements but sacrifice performance. If you do not have enough VRAM, web UI may refuse to launch or fail to generate images due to an out-of-memory error. The amount of VRAM required largely depends the desired image resolution, for more details see Troubleshooting.

the Tiled VAE extension can help to reduce the VRAM requirement.

  1. If your generated results in resulting in a black or green image, try adding --precision full and --no-half.

  2. Some combinations of model and VAE are prone to produce NansException: A tensor with all NaNs was produced in VAE resulting in a black image, using the option --no-half-vae may help to mitigate this issue.

Extra Options

  1. There are several cross attention optimization methods such as --xformers or --opt-sdp-attention, these can drastically increase performance see Optimizations for more details, experiment with different options as different hardware are suited for different optimizations. If you wish to measure your system's performance, try using sd-extension-system-info extension which features a benchmarking tool and a database of user submitted results.
  2. add --autolaunch to have web UI launch the web browser automatically after web UI has started up.
  3. add --update-check will notify you when there's a new version of webui.
  4. See Command Line Arguments and Settings for more configuration options.

Tip

If you already have stable diffusion models downloaded, you can move the models into sd.webui\webui\models\Stable-diffusion\ before running run.bat in step 3, this will skip auto downloading the vanilla stable-diffusion-v1-5 model model.

Windows (method 2)

  1. Install Python 3.10.6 (ticking Add to PATH), and git
  2. Open Command Prompt from search bar, and type git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
  3. Double click webui-user.bat

Installation video in case you get stuck:
solves #8229

Video: (Click to expand:)
webui_h264.mp4
Alternative Powershell launch scripts:

webui.ps1

if ($env:PYTHON -eq "" -or $env:PYTHON -eq $null) {
    $PYTHON = "Python.exe"
} else {
    $PYTHON = $env:PYTHON
}

if ($env:VENV_DIR -eq "" -or $env:VENV_DIR -eq $null) {
    $VENV_DIR = "$PSScriptRoot\venv"
} else {
    $VENV_DIR = $env:VENV_DIR
}

if ($env:LAUNCH_SCRIPT -eq "" -or $env:LAUNCH_SCRIPT -eq $null) {
    $LAUNCH_SCRIPT = "$PSScriptRoot\launch.py"
} else {
    $LAUNCH_SCRIPT = $env:LAUNCH_SCRIPT
}

$ERROR_REPORTING = $false

mkdir tmp 2>$null

function Start-Venv {
    if ($VENV_DIR -eq '-') {
        Skip-Venv
    }

    if (Test-Path -Path "$VENV_DIR\Scripts\$python") {
        Activate-Venv
    } else {
        $PYTHON_FULLNAME = & $PYTHON -c "import sys; print(sys.executable)"
        Write-Output "Creating venv in directory $VENV_DIR using python $PYTHON_FULLNAME"
        Invoke-Expression "$PYTHON_FULLNAME -m venv $VENV_DIR > tmp/stdout.txt 2> tmp/stderr.txt"
        if ($LASTEXITCODE -eq 0) {
            Activate-Venv
        } else {
            Write-Output "Unable to create venv in directory $VENV_DIR"
        }
    }
}

function Activate-Venv {
    $PYTHON = "$VENV_DIR\Scripts\Python.exe"
    $ACTIVATE = "$VENV_DIR\Scripts\activate.bat"
    Invoke-Expression "cmd.exe /c $ACTIVATE"
    Write-Output "Venv set to $VENV_DIR."
    if ($ACCELERATE -eq 'True') {
        Check-Accelerate
    } else {
        Launch-App
    }
}

function Skip-Venv {
    Write-Output "Venv set to $VENV_DIR."
    if ($ACCELERATE -eq 'True') {
        Check-Accelerate
    } else {
        Launch-App
    }
}

function Check-Accelerate {
    Write-Output 'Checking for accelerate'
    $ACCELERATE = "$VENV_DIR\Scripts\accelerate.exe"
    if (Test-Path -Path $ACCELERATE) {
        Accelerate-Launch
    } else {
        Launch-App
    }
}

function Launch-App {
    Write-Output "Launching with python"
    Invoke-Expression "$PYTHON $LAUNCH_SCRIPT"
    #pause
    exit
}

function Accelerate-Launch {
    Write-Output 'Accelerating'
    Invoke-Expression "$ACCELERATE launch --num_cpu_threads_per_process=6 $LAUNCH_SCRIPT"
    #pause
    exit
}


try {
    if(Get-Command $PYTHON){
        Start-Venv
    }
} Catch {
    Write-Output "Couldn't launch python."
}

webui-user.ps1

[Environment]::SetEnvironmentVariable("PYTHON", "")
[Environment]::SetEnvironmentVariable("GIT", "")
[Environment]::SetEnvironmentVariable("VENV_DIR","")

# Commandline arguments for webui.py, for example: [Environment]::SetEnvironmentVariable("COMMANDLINE_ARGS", "--medvram --opt-split-attention")
[Environment]::SetEnvironmentVariable("COMMANDLINE_ARGS", "")

# script to launch to start the app
# [Environment]::SetEnvironmentVariable("LAUNCH_SCRIPT", "launch.py")

# install command for torch
# [Environment]::SetEnvironmentVariable("TORCH_COMMAND", "pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 --extra-index-url https://download.pytorch.org/whl/cu113")

# Requirements file to use for stable-diffusion-webui
# [Environment]::SetEnvironmentVariable("REQS_FILE", "requirements_versions.txt")

# [Environment]::SetEnvironmentVariable("GFPGAN_PACKAGE", "")
# [Environment]::SetEnvironmentVariable("CLIP_PACKAGE", "")
# [Environment]::SetEnvironmentVariable("OPENCLIP_PACKAGE", "")

# URL to a WHL if you wish to override default xformers windows
# [Environment]::SetEnvironmentVariable("XFORMERS_WINDOWS_PACKAGE", "")

# Uncomment and set to enable an alternate repository URL
# [Environment]::SetEnvironmentVariable("STABLE_DIFFUSION_REPO", "")
# [Environment]::SetEnvironmentVariable("TAMING_TRANSFORMERS_REPO", "")
# [Environment]::SetEnvironmentVariable("K_DIFFUSION_REPO", "")
# [Environment]::SetEnvironmentVariable("CODEFORMER_REPO", "")
# [Environment]::SetEnvironmentVariable("BLIP_REPO", "")

# Uncomment and set to enable a specific revision of a repository
# [Environment]::SetEnvironmentVariable("STABLE_DIFFUSION_COMMIT_HASH", "")
# [Environment]::SetEnvironmentVariable("TAMING_TRANSFORMERS_COMMIT_HASH", "")
# [Environment]::SetEnvironmentVariable("K_DIFFUSION_COMMIT_HASH", "")
# [Environment]::SetEnvironmentVariable("CODEFORMER_COMMIT_HASH", "")
# [Environment]::SetEnvironmentVariable("BLIP_COMMIT_HASH", "")


# Uncomment to enable accelerated launch
# [Environment]::SetEnvironmentVariable("ACCELERATE", "True")

$SCRIPT = "$PSScriptRoot\webui.ps1"
Invoke-Expression "$SCRIPT"

See Troubleshooting section for what to do if things go wrong.

Linux

(Debian-based)

  1. Enter these commands, which will install webui to your current directory:
sudo apt install git python3.10-venv -y
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui && cd stable-diffusion-webui
python3.10 -m venv venv

  1. Install and run with:

    ./webui.sh {your_arguments}

Other Distributions:

(Arch-based)

sudo pacman -S git python3 -y && git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui && cd stable-diffusion-webui && ./webui.sh

(Red Hat-based)

sudo dnf install git python3 -y && git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui && cd stable-diffusion-webui && ./webui.sh
For Installing Python 3.10

Some distribution have older/newer system python versions than python 3.10.

cd stable-diffusion-webui

sudo pacman -S pyenv
pyenv install 3.10.6
pyenv local 3.10.6

python -m venv venv

Third party installation guides/scripts:

Install and run without virtual environment

To install the required packages via pip without creating a virtual environment, run:

python launch.py

Command line arguments may be passed directly, for example:

python launch.py --opt-split-attention --ckpt ../secret/anime9999.ckpt

Manual Installation

Manual installation is very outdated and probably won't work. check colab in the repo's readme for instructions.

The following process installs everything manually on both Windows or Linux (the latter requiring dir to be replaced by ls):

# install torch with CUDA support. See https://pytorch.org/get-started/locally/ for more instructions if this fails.
pip install torch --extra-index-url https://download.pytorch.org/whl/cu113

# check if torch supports GPU; this must output "True". You need CUDA 11. installed for this. You might be able to use
# a different version, but this is what I tested.
python -c "import torch; print(torch.cuda.is_available())"

# clone web ui and go into its directory
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
cd stable-diffusion-webui

# clone repositories for Stable Diffusion and (optionally) CodeFormer
mkdir repositories
git clone https://github.com/CompVis/stable-diffusion.git repositories/stable-diffusion-stability-ai
git clone https://github.com/CompVis/taming-transformers.git repositories/taming-transformers
git clone https://github.com/sczhou/CodeFormer.git repositories/CodeFormer
git clone https://github.com/salesforce/BLIP.git repositories/BLIP

# install requirements of Stable Diffusion
pip install transformers==4.19.2 diffusers invisible-watermark --prefer-binary

# install k-diffusion
pip install git+https://github.com/crowsonkb/k-diffusion.git --prefer-binary

# (optional) install GFPGAN (face restoration)
pip install git+https://github.com/TencentARC/GFPGAN.git --prefer-binary

# (optional) install requirements for CodeFormer (face restoration)
pip install -r repositories/CodeFormer/requirements.txt --prefer-binary

# install requirements of web ui
pip install -r requirements.txt  --prefer-binary

# update numpy to latest version
pip install -U numpy  --prefer-binary

# (outside of command line) put stable diffusion model into web ui directory
# the command below must output something like: 1 File(s) 4,265,380,512 bytes
dir model.ckpt

The installation is finished, to start the web ui, run:

python webui.py

Windows 11 WSL2 instructions

To install under a Linux distro in Windows 11's WSL2:

# install conda (if not already done)
wget https://repo.anaconda.com/archive/Anaconda3-2022.05-Linux-x86_64.sh
chmod +x Anaconda3-2022.05-Linux-x86_64.sh
./Anaconda3-2022.05-Linux-x86_64.sh

# Clone webui repo
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
cd stable-diffusion-webui

# Create and activate conda env
conda env create -f environment-wsl2.yaml
conda activate automatic

At this point, the instructions for the Manual installation may be applied starting at step # clone repositories for Stable Diffusion and (optionally) CodeFormer.

Alternative installation on Windows using Conda

  • Prerequisites (Only needed if you do not have them). Assumes Chocolatey is installed. 
    # install git
choco install git
# install conda
choco install anaconda3
    Optional parameters: git, conda
  • Install (warning: some files exceed multiple gigabytes, make sure you have space first)
    1. Download as .zip and extract or use git to clone. 
      git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git
    2. Launch the Anaconda prompt. It should be noted that you can use older Python versions, but you may be forced to manually remove features like cache optimization, which will degrade your performance. 
      # Navigate to the git directory
cd "GIT\StableDiffusion"
# Create environment
conda create -n StableDiffusion python=3.10.6
# Activate environment
conda activate StableDiffusion
# Validate environment is selected
conda env list
# Start local webserver
webui-user.bat
# Wait for "Running on local URL:  http://127.0.0.1:7860" and open that URI.
    3. (Optional) Go to CompVis and download latest model, for example 1.4 and unpack it to ex: 
      GIT\StableDiffusion\models\Stable-diffusion
      after that restart the server by restarting Anaconda prompt and 
      webui-user.bat
  • Alternative defaults worth trying out:
    1. Try euler a (Ancestral Euler) with higher Sampling Steps ex: 40 or others with 100.
    2. Set "Settings > User interface > Show image creation progress every N sampling steps" to 1 and pick a deterministic Seed value. Can visually see how image diffusion happens and record a .gif with ScreenToGif.
    3. Use Restore faces. Generally, better results, but that quality comes at the cost of speed.

This is the Stable Diffusion web UI wiki. Wiki Home

Setup

Reproducing images / troubleshooting

Usage

Developers

Clone this wiki locally