Live stream set for 2025-10-01 at 14:00:00 Eastern
Ask questions in the live chat about any programming or lifestyle topic.
This livestream will be on YouTube or you can watch below.
Beginner’s Guide to Running ComfyUI on Fedora 42 with AMD Instinct MI60
If you have ever wanted to generate AI images locally without relying on expensive cloud services, ComfyUI is a powerful open-source tool that makes it easy — even on non-NVIDIA GPUs like the AMD Instinct MI60.
In this beginner-friendly tutorial, I will show you how to get ComfyUI up and running on Fedora Linux 42 with an AMD Instinct MI60 32GB GPU. This setup leverages the ROCm stack to unlock the GPU’s full performance – all while keeping your workflow fully local and private.
What is ComfyUI?
ComfyUI is a node-based, open-source interface for Stable Diffusion models. It offers:
- A visual, modular workflow builder
- Support for ControlNet, LoRAs, VAEs, and more
- Performance optimizations
- Community-developed custom nodes
Unlike some web UIs that rely heavily on CUDA/NVIDIA, ComfyUI’s modular structure makes it more adaptable to AMD GPUs via PyTorch with ROCm.
System Used in This Tutorial
- GPU: AMD Instinct MI60 (32GB HBM2)
- OS: Fedora Linux 42 (Workstation Edition)
- Tool: ComfyUI (Stable Diffusion interface)
- Framework: PyTorch (ROCm build)
- License: Fully Open Source
Installation Guide – ComfyUI on Fedora 42 with AMD MI60
You can follow these steps to get a working ComfyUI environment using AMD ROCm and PyTorch ROCm builds.
Step 1: Install System Dependencies
sudo dnf install git python3 python3-virtualenv rocm-hip rocm-libs
Then check if your GPU is detected:
rocminfo
rocm-smi --showmeminfo vram
Step 2: Clone ComfyUI
git clone https://github.com/comfyanonymous/ComfyUI.git
cd ComfyUI
Step 3: Create a Python Virtual Environment
python3 -m venv venv
source venv/bin/activate
Step 4: Install PyTorch (ROCm Build) and Torchaudio
pip install torch==2.2.0+rocm5.7 torchaudio torchvision --extra-index-url https://download.pytorch.org/whl/rocm5.7
Step 5: Install ComfyUI Dependencies
pip install -r requirements.txt
Step 6: Run ComfyUI
python3 main.py
Open your browser and go to:
http://localhost:8188
Where to Place Your Models
| Model Type | Folder |
|---|---|
| Checkpoints (e.g. *.safetensors) | ComfyUI/models/checkpoints/ |
| LoRA models | ComfyUI/models/loras/ |
| VAEs | ComfyUI/models/vae/ |
| ControlNet | ComfyUI/models/controlnet/ |
Just drop your files into the appropriate folder and restart ComfyUI.
Test Tools
| Name | Description |
|---|---|
| CPU | AMD Ryzen 5 5600GT (6C/12T, 3.6GHz). |
| Memory | 32GB DDR4. |
| GPU | AMD Instinct MI60 (32GB HBM2). |
| Operating System | Fedora Linux Workstation 42. |
| Desktop Environment | Gnome 48. |
| Name | Description |
🖼️ Screenshots and Screencast
Results:
A photograph of the mayor of Toronto
Accurately drew a photograph mishmash of past mayors of Toronto.
A screenshot of the gnome desktop environment.
Accurately drew a screenshot of an older version of the Gnome desktop environment.
A photograph of an astronaut riding a horse.
Accurately drew a photograph of an astronaut riding a horse.
A picture of a chicken run.
Accurately drew a picture of a chicken run.
A picture of a man wearing a watch.
Accurately drew a picture of a man wearing a watch.
A picture of a spider web on sockets.
Accurately drew a picture of a spider web on sockets.
Learn More About Python
If you are curious about how AI tools like ComfyUI are built, or you want to automate your own workflows, learning Python is the perfect next step.
Book: Learning Python
This beginner-friendly guide walks you through the fundamentals of Python programming with clear examples.
Available on Amazon
Course: Learning Python
Prefer hands-on learning? Take my interactive Python course.
Get the course at OJambo Shop
1-on-1 Python Tutoring
Want help understanding ComfyUI workflows, or need personalized Python training?
Book a private online session
Final Thoughts
Running ComfyUI on Fedora 42 with an AMD Instinct MI60 is absolutely possible – and powerful. With open-source tools like ComfyUI and ROCm, you no longer need to rely on cloud APIs or NVIDIA hardware to start generating stunning AI images.
Have questions or want help with installation? Feel free to reach out.
Disclosure: Some of the links above are referral (affiliate) links. I may earn a commission if you purchase through them - at no extra cost to you.
Edward is a software engineer, web developer, and author dedicated to helping people achieve their personal and professional goals through actionable advice and real-world tools.