Skip to content

Latest commit

 

History

History
141 lines (120 loc) · 5.43 KB

installing.md

File metadata and controls

141 lines (120 loc) · 5.43 KB
Raw

Install Automatic1111 Web UI on your Network Volume

Note

If you have previously installed A1111 on your Network volume, but want to install the After Detailer extension, you can follow the instructions here.

  1. Create a RunPod Account.
  2. Create a RunPod Network Volume.
  3. Attach the Network Volume to a Secure Cloud GPU pod.
  4. Select a light-weight template such as RunPod Pytorch.
  5. Deploy the GPU Cloud pod.
  6. Once the pod is up, open a Terminal and install the required dependencies. This can either be done by using the installation script, or manually.

Automatic Installation Script

You can run this automatic installation script which will automatically install all of the dependencies that get installed manually below, and then you don't need to follow any of the manual instructions.

wget https://raw.githubusercontent.com/ashleykleynhans/runpod-worker-a1111/main/scripts/install.sh
chmod +x install.sh
./install.sh

Manual Installation

You only need to complete the steps below if you did not run the automatic installation script above.

  1. Install the Automatic1111 WebUI and ControlNet extension:
# Clone the repo
cd /workspace
git clone --depth=1 https://github.com/AUTOMATIC1111/stable-diffusion-webui.git

# Upgrade Python
apt update
apt -y upgrade

# Ensure Python version is 3.10.12
python3 -V

# Create and activate venv
cd stable-diffusion-webui
python3 -m venv /workspace/venv
source /workspace/venv/bin/activate

# Install Torch and xformers
pip3 install --no-cache-dir torch==2.0.1+cu118 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
pip3 install --no-cache-dir xformers==0.0.22

# Install A1111 Web UI
wget https://raw.githubusercontent.com/ashleykleynhans/runpod-worker-a1111/main/install-automatic.py
python3 -m install-automatic --skip-torch-cuda-test

# Clone the ControlNet Extension
cd /workspace/stable-diffusion-webui
git clone --depth=1 https://github.com/Mikubill/sd-webui-controlnet.git extensions/sd-webui-controlnet

# Clone the ReActor Extension
git clone --depth=1 https://github.com/Gourieff/sd-webui-reactor.git extensions/sd-webui-reactor

# Clone the After Detailer Extension
git clone --depth=1 https://github.com/Bing-su/adetailer.git extensions/adetailer

# Install dependencies for ControlNet
cd /workspace/stable-diffusion-webui/extensions/sd-webui-controlnet
pip3 install -r requirements.txt

# Install dependencies for ReActor
cd /workspace/stable-diffusion-webui/extensions/sd-webui-reactor
pip3 install -r requirements.txt
pip3 install onnxruntime-gpu

# Install dependencies for After Detailer
cd /workspace/stable-diffusion-webui/extensions/adetailer
python3 -m install

# Install the model for ReActor
mkdir -p /workspace/stable-diffusion-webui/models/insightface
cd /workspace/stable-diffusion-webui/models/insightface
wget https://github.com/facefusion/facefusion-assets/releases/download/models/inswapper_128.onnx

# Configure ReActor to use the GPU instead of CPU
echo "CUDA" > /workspace/stable-diffusion-webui/extensions/sd-webui-reactor/last_device.txt
  1. Install the Serverless dependencies:
cd /workspace/stable-diffusion-webui
pip3 install huggingface_hub runpod
  1. Download some models, for example SDXL and Deliberate v2:
cd /workspace/stable-diffusion-webui/models/Stable-diffusion
wget https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors
wget https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-1.0/resolve/main/sd_xl_refiner_1.0.safetensors
wget -O deliberate_v2.safetensors https://huggingface.co/XpucT/Deliberate/resolve/main/Deliberate_v2.safetensors
  1. Download VAEs for SD 1.5 and SDXL:
cd /workspace/stable-diffusion-webui/models/VAE
wget https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.safetensors
wget https://huggingface.co/madebyollin/sdxl-vae-fp16-fix/resolve/main/sdxl_vae.safetensors
  1. Download ControlNet models, for example canny for SD 1.5 as well as SDXL:
mkdir -p /workspace/stable-diffusion-webui/models/ControlNet
cd /workspace/stable-diffusion-webui/models/ControlNet
wget https://huggingface.co/lllyasviel/ControlNet-v1-1/resolve/main/control_v11p_sd15_canny.pth
wget https://huggingface.co/lllyasviel/sd_control_collection/resolve/main/diffusers_xl_canny_full.safetensors
  1. Create logs directory:
mkdir -p /workspace/logs
  1. Install config files:
cd /workspace/stable-diffusion-webui
rm webui-user.sh config.json ui-config.json
wget https://raw.githubusercontent.com/ashleykleynhans/runpod-worker-a1111/main/webui-user.sh
wget https://raw.githubusercontent.com/ashleykleynhans/runpod-worker-a1111/main/config.json
wget https://raw.githubusercontent.com/ashleykleynhans/runpod-worker-a1111/main/ui-config.json
  1. Run the Web UI:
deactivate
export HF_HOME="/workspace"
cd /workspace/stable-diffusion-webui
./webui.sh -f
  1. Wait for the Web UI to start up, and download the models. You shoud see something like this when it is ready:
Model loaded in 16.9s (calculate hash: 8.0s, load weights from disk: 0.4s, create model: 2.1s, apply weights to model: 2.6s, apply half(): 2.6s, move model to device: 0.7s, calculate empty prompt: 0.3s).
  1. Press Ctrl-C to exit, and then you can terminate the pod.