This repo is for Docker images that runs ComfyUI - a Stable Diffusion GUI powering node-based workflow.
-
At first start, a script will download latest ComfyUI, ComfyUI-Manager and some models.
-
The whole ComfyUI will be stored in a local folder (
./storage/ComfyUI). -
If you already have a ComfyUI bundle, put it there and make an empty file (
./storage/.download-complete) so the start script will skip downloading. -
Use ComfyUI-Manager (in ComfyUI web page) to update ComfyUI, manage custom nodes, and download models.
-
NVIDIA GPU with ≥6GB VRAM
-
Latest NVIDIA GPU driver
-
Either Game or Studio edition will work.
-
You don’t need to install drivers inside containers. Just make sure it’s working on your host OS.
-
-
Docker Installed
-
Linux user may need to install NVIDIA Container Toolkit (only on host OS). It will enable containers' GPU access.
-
Windows user could use Docker Desktop with WSL2 enabled.
-
For Podman users see: Run with Podman.
-
docker runmkdir -p storage
docker run -it \
--name comfyui \
--gpus all \
-p 8188:8188 \
-v "$(pwd)"/storage:/home/runner \
-e CLI_ARGS="" \
yanwk/comfyui-boot:latest# Update image (only when Python components are outdated)
docker rm comfyui
docker pull yanwk/comfyui-boot:latest
# Then re-run above 'docker run' againdocker composegit clone https://github.com/YanWenKun/ComfyUI-Docker.git
cd ComfyUI-Docker
docker compose up --detach# Update image (only when Python components are outdated)
git pull
docker compose pull
docker compose up --detach --remove-orphans
docker image pruneOnce the app is loaded, visit http://localhost:8188/
See ROCm.
Q: My GPU has only 4GB VRAM
A: Add --lowvram to CLI_ARGS.
Q: Adding --lowvram still out-of-memory
A: Use --novram instead. It will use CPU RAM.
Q: How to run on CPU?
A: Add --cpu to CLI_ARGS. It’s pretty slow.
Q: I don’t want to use xFormers, how to use PyTorch’s Cross Attention?
A: Add --use-pytorch-cross-attention to CLI_ARGS. It may perform well on WSL2, but significantly slower on Linux hosts.
More CLI_ARGS available at
ComfyUI.
Install a bunch of custom nodes that may save your time
Note that most dependencies are bundled in the image, you don’t need to manually install them.
cd ComfyUI/custom_nodes/
gcs='git clone --depth=1 --no-tags --recurse-submodules --shallow-submodules'
# Workspace
$gcs https://github.com/11cafe/comfyui-workspace-manager.git
$gcs https://github.com/AIGODLIKE/AIGODLIKE-ComfyUI-Translation.git
$gcs https://github.com/crystian/ComfyUI-Crystools-save.git
$gcs https://github.com/crystian/ComfyUI-Crystools.git
# General
$gcs https://github.com/bash-j/mikey_nodes.git
$gcs https://github.com/chrisgoringe/cg-use-everywhere.git
$gcs https://github.com/cubiq/ComfyUI_essentials.git
$gcs https://github.com/Derfuu/Derfuu_ComfyUI_ModdedNodes.git
$gcs https://github.com/jags111/efficiency-nodes-comfyui.git
$gcs https://github.com/kijai/ComfyUI-KJNodes.git
$gcs https://github.com/pythongosssss/ComfyUI-Custom-Scripts.git
$gcs https://github.com/rgthree/rgthree-comfy.git
$gcs https://github.com/shiimizu/ComfyUI_smZNodes.git
$gcs https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes.git
# Control
$gcs https://github.com/cubiq/ComfyUI_InstantID.git
$gcs https://github.com/cubiq/ComfyUI_IPAdapter_plus.git
$gcs https://github.com/Fannovel16/comfyui_controlnet_aux.git
$gcs https://github.com/florestefano1975/comfyui-portrait-master.git
$gcs https://github.com/Gourieff/comfyui-reactor-node.git
$gcs https://github.com/huchenlei/ComfyUI-layerdiffuse.git
$gcs https://github.com/Kosinkadink/ComfyUI-Advanced-ControlNet.git
$gcs https://github.com/ltdrdata/ComfyUI-Impact-Pack.git
$gcs https://github.com/ltdrdata/ComfyUI-Inspire-Pack.git
$gcs https://github.com/mcmonkeyprojects/sd-dynamic-thresholding.git
$gcs https://github.com/storyicon/comfyui_segment_anything.git
$gcs https://github.com/twri/sdxl_prompt_styler.git
# Video
$gcs https://github.com/Fannovel16/ComfyUI-Frame-Interpolation.git
$gcs https://github.com/FizzleDorf/ComfyUI_FizzNodes.git
$gcs https://github.com/Kosinkadink/ComfyUI-AnimateDiff-Evolved.git
$gcs https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite.git
$gcs https://github.com/melMass/comfy_mtb.git
$gcs https://github.com/MrForExample/ComfyUI-AnimateAnyone-Evolved.git
# More
$gcs https://github.com/cubiq/ComfyUI_FaceAnalysis.git
$gcs https://github.com/MrForExample/ComfyUI-3D-Pack.git
$gcs https://github.com/pythongosssss/ComfyUI-WD14-Tagger.git
$gcs https://github.com/SLAPaper/ComfyUI-Image-Selector.git
$gcs https://github.com/ssitu/ComfyUI_UltimateSDUpscale.gitAnother note is this image doesn’t bundled deps for
WAS Node Suite,
because it has some deps version
fixed,
and is not under active development.
However, the image’s script only install one custom node (ComfyUI-Manager). You can safely install WAS NS via ComfyUI-Manager in a new deployment.
If anything conflicts, just delete unwanted custom nodes and .local (or local) folder, and update/try-fix/reinstall custom nodes in ComfyUI-Manager.
You can create your own pre-start script at
./storage/scripts/pre-start.sh
It will run before ComfyUI starts, handy for debugging.
Also, if you need to setup proxy, this script will start before pre-start and download :
./storage/scripts/set-proxy.sh
Podman is root-less by default. And it’s tricky to bind-mount volume as non-root user inside rootless container. Options provided by Podman will chown files on host space, which probably is undesirable.
Recommend three different ways to workaround:
Expand details
The straightforward way. By adding sudo you go from rootless to rootful. And everything else would be the same as using Docker.
Note that sudo Podman will download images to root space. If you already downloaded the image in current user, you can local-copy it:
sudo podman image scp username@localhost::docker.io/yanwk/comfyui-boot:latest
mkdir -p storage
sudo podman run -it --rm \
--name comfyui-rootful \
--device nvidia.com/gpu=all \
--security-opt label=disable \
-p 8188:8188 \
-v "$(pwd)"/storage:/home/runner \
-e CLI_ARGS="" \
docker.io/yanwk/comfyui-bootExpand details
The rootless way, no sudo needed. Inside the pod, Podman will mount volume as root, scripts will run as root. And from the host side, we see files keep their original ownership.
mkdir -p storage
podman run -it --rm \
--name comfyui-rootless \
--device nvidia.com/gpu=all \
--security-opt label=disable \
-p 8188:8188 \
-v "$(pwd)"/storage:/root \
--user root \
--workdir /root \
-e CLI_ARGS="" \
docker.io/yanwk/comfyui-boot:latest \
/bin/bash /home/scripts/root-wrapper.sh3. Use megapak image
This image was built for rootless.
docker build . --progress=plain -f Dockerfile -t yanwk/comfyui-boot:latestdocker run -it --rm \
--gpus all -p 8188:8188 \
--volume "$(pwd)"/storage:/home/runner \
--env CLI_ARGS="" \
yanwk/comfyui-boot:latestdocker run -it --rm \
--gpus all -p 8188:8188 \
--volume "$(pwd)"/storage:/home/runner \
--env CLI_ARGS="" \
--user root \
yanwk/comfyui-boot:latest /bin/bashUsually this is not needed.
Some custom nodes will use huggingface_hub to download models and store them in .cache.
You may need to download them again after cleaning cache.
But if having issues when updating, it may worth a try:
docker exec -it --workdir /home/runner comfyui \
rm -rf .cache/ .config/ .local/ .nv/ bin/ include/ lib/ lib64 pyvenv.cfg
docker restart comfyuiThis open source license is written and valid both in Chinese and English, how good is that!
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent