Skip to content

Latest commit

 

History

History
86 lines (58 loc) · 3.24 KB

DOCKER_INSTRUCTIONS.md

File metadata and controls

86 lines (58 loc) · 3.24 KB
Raw

Docker Instructions

This document contains instructions on how to build and run the application using Docker.

Disclaimer

These instructions are based on the assumption that you have Docker installed on your machine. If you don't have Docker installed, you can download it from the official website.

All tests were performed using Docker version 26.1.3 and Docker Compose version 2.27.0 under Ubuntu 22.04.

These instructions are focused on running the Unique3D workflows, you can use the same steps to run the other workflows, but you need to check the model requirements for each workflow at the README.md file.

Requirements

You need the nvidia-container-runtime installed and configured to run the application. You can check the installation instructions at the official documentation.

Models

To use Unique3D you will need to download the following models and place them at the models folder in the root directory:

wget -c https://huggingface.co/h94/IP-Adapter/resolve/main/models/ip-adapter_sd15.safetensors -P models/
wget -c https://huggingface.co/h94/IP-Adapter/resolve/main/models/image_encoder/model.safetensors -P models/
wget -c https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth -P models/
wget -c https://huggingface.co/spaces/Wuvin/Unique3D/resolve/main/ckpt/controlnet-tile/diffusion_pytorch_model.safetensors -P models/
wget -c https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.ckpt -P models/

You can check more details at the Unique3D section of the README.md file.

Build Docker Image

You can build the docker image using the following command:

DOCKER_BUILDKIT=0 docker build -t comfy3d --file Dockerfile .

You can also build the image using the docker-compose file:

DOCKER_BUILDKIT=0 docker compose build

OBS: The DOCKER_BUILDKIT=0 is used to disable the buildkit feature, which is not supported by the docker-compose file.

Run Docker Container

You can run the docker container using the following command:

docker run -d \
  --name comfy3d \
  --platform linux/amd64 \
  --runtime=nvidia \
  -p 8188:8188 \
  -e NVIDIA_VISIBLE_DEVICES=all \
  -e NVIDIA_DRIVER_CAPABILITIES=all \
  -e COMFYUI_PATH=/app \
  -v $(pwd)/output:/app/output \
  -v $(pwd)/models/diffusion_pytorch_model.safetensors:/app/models/controlnet/control_unique3d_sd15_tile.safetensors \
  -v $(pwd)/models/ip-adapter_sd15.safetensors:/app/models/ipadapter/ip-adapter_sd15.safetensors \
  -v $(pwd)/models/RealESRGAN_x4plus.pth:/app/models/upscale_models/RealESRGAN_x4plus.pth \
  -v $(pwd)/models/v1-5-pruned-emaonly.ckpt:/app/models/checkpoints/v1-5-pruned-emaonly.ckpt \
  -v $(pwd)/models/model.safetensors:/app/models/clip_vision/OpenCLIP-ViT-H-14.safetensors \
  comfy3d

You can also run the container using the docker-compose file:

docker compose up

Access the Application

After running the container, you can access the application by opening the following URL in your browser:

http://localhost:8188

To test, you can load the following workflow:

  • ./_Example_Workflows/Unique3D/Unique3D_All_Stages.json