Docker image for SUPIR (Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild)

Important

This needs at least 24GB VRAM for 1x upscale. If you want to upscale more than 1x, you will need more than 24GB of VRAM. 48GB VRAM is recommended.

Note

Loading of models on start takes a few minutes, so you can view the log to watch the progress. You will be able to access the port when you see Running on local URL: http://0.0.0.0:3001 in the log.

Installs

Ubuntu 22.04 LTS
CUDA 12.1
Python 3.10.12
SUPIR
Torch 2.3.1
xformers 0.0.27
Jupyter Lab
code-server
runpodctl
OhMyRunPod
RunPod File Uploader
croc
rclone

Available on RunPod

This image is designed to work on RunPod. You can use my custom RunPod template to launch it on RunPod.

Building the Docker image

Note

You will need to edit the docker-bake.hcl file and update USERNAME, and RELEASE. You can obviously edit the other values too, but these are the most important ones.

# Clone the repo
git clone https://github.com/ashleykleynhans/supir-docker.git

# Log in to Docker Hub
docker login

# Build the image, tag the image, and push the image to Docker Hub
cd supir-docker
docker buildx bake -f docker-bake.hcl --push

Running Locally

Install Nvidia CUDA Driver

Start the Docker container

docker run -d \
  --gpus all \
  -v /workspace \
  -p 3000:3001 \
  -p 7777:7777 \
  -p 8888:8888 \
  -p 2999:2999 \
  ashleykza/supir:latest

Models

Model	Description
SUPIR-v0F.ckpt	SUPIR F
SUPIR-v0Q.ckpt	SUPIR Q
liuhaotian/llava-v1.5-7b	LLaVA
sd_xl_base_1.0_0.9vae.safetensors	SDXL
openai/clip-vit-large-patch14-336	LLaVA CLIP
openai/clip-vit-large-patch14	SDXL CLIP1
open_clip_pytorch_model.bin	SDXL CLIP2

You can obviously substitute the image name and tag with your own.

Ports

Connect Port	Internal Port	Description
3000	3001	SUPIR
7777	7777	Code Server
8888	8888	Jupyter Lab
2999	2999	RunPod File Uploader

Environment Variables

Variable	Description	Default
JUPYTER_LAB_PASSWORD	Set a password for Jupyter lab	not set - no password
DISABLE_AUTOLAUNCH	Disable SUPIR from launching automatically	(not set)
DISABLE_SYNC	Disable syncing if using a RunPod network volume	(not set)
NO_GPU_OPTIMIZATION	Disable GPU optimization for A100/H100	(not set)

Logs

SUPIR creates a log file, and you can tail the log instead of killing the service to view the logs.

Application	Log file
SUPIR	/workspace/logs/supir.log

For example:

tail -f /workspace/logs/supir.log

Community and Contributing

Pull requests and issues on GitHub are welcome. Bug fixes and new features are encouraged.

ashleykleynhans/supir-docker