OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Comfyui Docker

Comfyui Docker is a GPL-3.0-licensed Docker image that runs ComfyUI, a node-and-graph Stable Diffusion interface, API and backend.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Comfyui Docker — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a Docker-packaged ComfyUI environment, need to build a variant, or are selecting one of the repository’s linked RunPod templates.

Who it is for

  • People packaging ComfyUI as a Docker image for image-generation workflows.
  • RunPod users choosing among the repository’s linked CUDA and Python template combinations.

Documented uses

  • Build and push a ComfyUI image variant with Docker Buildx Bake.
  • Select a documented CUDA and Python combination for a target RunPod GPU type.
  • Run ComfyUI workflows through its graph-based interface and backend.
Before you choose

Requirements & limits

  • The listed CUDA 12.4 templates are for non-RTX 5090 GPU types; CUDA 12.8 templates are listed for RTX 5090.
  • Python 3.11 may suit some custom nodes that the README says do not work correctly with Python 3.12.
  • Model caching requires at least 32 GB of system memory according to the README.
  • The image includes JupyterLab, code-server and RunPod-related utilities in addition to ComfyUI.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Build and push a ComfyUI Docker image

Build and push a selected image variant using the repository’s documented Docker Buildx Bake process.

Before you start

  • Docker and Docker Buildx
  • A cloned copy of the repository
  • Registry and release values set in docker-bake.hcl
  • At least 32 GB of system memory if retaining model caching

Steps

  1. Clone the repository.

  2. Edit docker-bake.hcl to update REGISTRY_USER and RELEASE.

  3. Log in to Docker Hub.

  4. Run the documented Docker Buildx Bake command for the default image, a named variant, or all images, using --push.

Expected outcome

A selected ComfyUI Docker image is built and pushed to the configured registry.

Guide limitations

  • The README says model caching requires at least 32 GB of system memory; with less, remove or comment out the model-caching code in the Dockerfile.
  • The documented commands do not demonstrate a successful build or push.
Documentation 1 ↗
Source-based practical guide

Choose a linked RunPod template variant

Select a RunPod template matching the documented CUDA and Python combination.

Before you start

  • A choice of GPU type and Python version
  • Access to the linked RunPod template options

Steps

  1. Choose CUDA 12.4 for a non-RTX 5090 GPU type or CUDA 12.8 for an RTX 5090 GPU type.

  2. Choose Python 3.11 or 3.12, accounting for the README’s note that some custom nodes do not work correctly with Python 3.12.

  3. Open the corresponding RunPod template linked from the repository.

Expected outcome

A matching ComfyUI template is selected for the documented GPU, CUDA and Python combination.

Guide limitations

  • The repository links the templates but does not provide their deployment configuration or demonstrate a successful deployment.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
ashleykleynhans/comfyui-docker · Official homepage title · retained in its source language
How we review sourcesSuggest a correction