[01 / Pick Your Path]

Choose your install.

Four ways to run ComfyUI. There's no "right" one — pick the one that matches your hardware. If you have no GPU and no patience, skip straight to Google Colab.

Before you begin
Minimum specs
for local install.
→ GPU VRAM
6 GB minimum
12 GB ideal
→ System RAM
16 GB minimum
32 GB ideal
→ Disk Space
40 GB free
(models are big)
→ Internet
~5 GB download
first time
▸ Path A01

Windows · portable install.

The official ComfyUI Windows portable package — no Python setup, no environment hassle. Download, unzip, double-click. This is the fastest path.

Step 1 · Check your GPU

Press Win + R, type dxdiag, press Enter. Click the "Display" tab. Read what's next to "Name". You're looking for an NVIDIA GeForce card.

  • NVIDIA RTX 30/40/50-series — excellent, will run everything in this workshop
  • NVIDIA RTX 20-series or GTX 16/10-series with ≥6GB — works for most things, slower on Flux/SDXL
  • AMD Radeon — possible but painful on Windows. Use Colab instead
  • Intel integrated / no GPU — skip to Google Colab

Step 2 · Download ComfyUI portable

Open github.com/comfyanonymous/ComfyUI/releases in a browser. Under the latest release, find the file ending in _nvidia.7z (it's around 2 GB).

→ The file you want

Look for something like ComfyUI_windows_portable_nvidia.7z. Avoid any file with "directml" or "cpu" in the name — those are for AMD or CPU-only setups, which are slow.

Step 3 · Extract with 7-Zip

You'll need 7-Zip to extract the file. Download it from 7-zip.org if you don't have it. Right-click the downloaded archive → 7-Zip → Extract Here.

This creates a folder called ComfyUI_windows_portable. Important: put this folder somewhere with at least 40 GB of free space — your C:\ drive root, or a fast external SSD.

Step 4 · First launch

Inside the extracted folder you'll see two batch files:

  • run_nvidia_gpu.bat — use this one
  • run_cpu.bat — only if you don't have NVIDIA

Double-click run_nvidia_gpu.bat. A black command window opens. After about 30 seconds your default browser opens to 127.0.0.1:8188 with the ComfyUI interface.

cmd.exe — first run
# What you'll see on first launch: [ComfyUI] Starting server... [ComfyUI] Total VRAM 12288 MB, total RAM 32626 MB [ComfyUI] Set vram state to: NORMAL_VRAM [ComfyUI] Device: cuda:0 NVIDIA GeForce RTX 4070 [ComfyUI] Using pytorch attention [ComfyUI] To see the GUI go to: http://127.0.0.1:8188
→ You're in

If the browser opened and you see a node graph with a few connected boxes, ComfyUI is installed. You can try clicking Queue Prompt right now — but you'll get an error because no model is loaded yet. That's next.

Step 5 · Download your first model

ComfyUI by itself can't generate anything — it needs a model (also called a "checkpoint"). For this workshop, start with one of these:

Place the downloaded .safetensors file inside: ComfyUI_windows_portable\ComfyUI\models\checkpoints\

Step 6 · Load model and generate

Back in the ComfyUI browser tab, click Refresh (top right). In the node called "Load Checkpoint", click the dropdown — your model should appear. Select it. Click Queue Prompt. Wait 10–30 seconds. You'll see your first AI image.

→ Win condition

If you see an image on the right side of the screen, your installation works. You're ready for Workshop Session 1.

▸ Path B02

macOS · Apple Silicon.

M1, M2, M3, M4 Macs run ComfyUI well using Apple's Metal acceleration. The install needs Terminal but the steps are copy-paste.

Step 1 · Install Homebrew (if you don't have it)

Open Terminal (press Cmd + Space, type Terminal, hit Enter). Paste this entire line and press Enter:

Terminal — install Homebrew
$ /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

It'll ask for your Mac password. Type it (you won't see characters, that's normal). Wait 5–10 minutes. When it finishes, run the two lines it tells you to run (about adding Homebrew to your PATH).

Step 2 · Install Python 3.11 and git

Terminal — install dependencies
$ brew install python@3.11 git

Step 3 · Clone ComfyUI

Terminal — clone repo
$ cd ~ $ git clone https://github.com/comfyanonymous/ComfyUI.git $ cd ComfyUI

Step 4 · Create a virtual environment and install

Terminal — Python setup
$ python3.11 -m venv venv $ source venv/bin/activate $ pip install --upgrade pip $ pip install torch torchvision torchaudio $ pip install -r requirements.txt

That last line takes about 5 minutes. Don't worry about the long output.

Step 5 · Launch

Terminal — run ComfyUI
$ python main.py To see the GUI go to: http://127.0.0.1:8188

Open that URL in Safari or Chrome. Same interface as Windows.

→ Each time you launch

Every new session, open Terminal and run: cd ~/ComfyUI && source venv/bin/activate && python main.py. Or save those three lines as a shell script you double-click.

Step 6 · Download models

Same as the Windows section — place .safetensors files inside ~/ComfyUI/models/checkpoints/.

→ Performance note

Apple Silicon is good, not great. Generation is roughly 3-5× slower than a similarly priced NVIDIA card. For Flux models specifically, you'll want a 16 GB or 24 GB unified memory Mac. If you have an 8 GB M-series, stick with SD 1.5 or SDXL Turbo.

▸ Path C03

Linux · Ubuntu setup.

If you're on Linux you probably know what you're doing. Here's the short version for Ubuntu 22.04+ with an NVIDIA GPU.

Pre-requisites

  • NVIDIA drivers installed: nvidia-smi should return your GPU
  • CUDA toolkit 12.x
  • Python 3.11 and git

Install

terminal — install
$ sudo apt update && sudo apt install -y python3.11 python3.11-venv git $ git clone https://github.com/comfyanonymous/ComfyUI.git ~/ComfyUI $ cd ~/ComfyUI $ python3.11 -m venv venv $ source venv/bin/activate $ pip install --upgrade pip $ pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121 $ pip install -r requirements.txt

Launch

terminal — launch
$ python main.py --listen 0.0.0.0

The --listen 0.0.0.0 flag exposes ComfyUI on your network — useful for headless Linux boxes. Access it from another machine at http://YOUR-LINUX-IP:8188. If using only locally, omit the flag.

▸ Path D · ★04

Google Colab · no GPU needed.

Run ComfyUI on a free Google GPU from any laptop — even a 10-year-old one. This is the path for anyone without an NVIDIA card. Setup is genuinely two clicks.

Step 1 · Open the workshop Colab notebook

We've prepared a Colab notebook with ComfyUI and a base model pre-configured. Click the link below to open it.

Step 2 · Enable GPU runtime

  1. In the notebook, click Runtime in the top menu
  2. Click Change runtime type
  3. Under "Hardware accelerator" choose T4 GPU
  4. Click Save

Step 3 · Run the cells

Click the ▶ play button on each cell from top to bottom. The first cell installs ComfyUI (~2 min), the second downloads a model (~2 min), the third launches the server.

At the end, you'll see a URL like https://random-words.trycloudflare.com. Click it. ComfyUI opens in a new tab, running on Google's GPU.

→ Best of both worlds

You get a free NVIDIA T4 (better than most laptops), no local install, and you can save your workflows to Google Drive. The catch: free tier disconnects after ~4 hours of inactivity. Plenty for any single workshop session.

→ Honest caveats

Free Colab is metered. Heavy daily use will eventually rate-limit you. For ₹800/month Colab Pro gives you longer sessions and better GPUs. Worth it if you start using ComfyUI seriously.

▸ Final05

After install — essential add-ons.

Three things to set up before you do anything else. Together they take 10 minutes and unlock 80% of what makes ComfyUI useful for architecture.

1 · ComfyUI Manager

The most important add-on. Lets you install everything else from within the UI instead of via command line.

Install method:

  • Open your ComfyUI/custom_nodes/ folder
  • Open a terminal there and run: git clone https://github.com/ltdrdata/ComfyUI-Manager.git
  • Restart ComfyUI. You'll see a new "Manager" button in the menu.

2 · ControlNet models

ControlNet is what makes ComfyUI useful for architects — it lets you feed in your linework, depth maps, or CAD exports. Install via the Manager:

  1. Click Manager → Install Models
  2. Search for controlnet
  3. Install at least: ControlNet Canny, ControlNet Depth, ControlNet Lineart

3 · A couple of LoRAs

LoRAs are small style add-ons (typically 50-300 MB) that fine-tune the model toward a specific aesthetic. For architecture, two are worth starting with:

  • Architecture Photography LoRA — pushes outputs toward editorial photo style
  • Concept Sketch LoRA — useful early-stage design exploration

Search for "architecture" on civitai.com. Download to ComfyUI/models/loras/.

→ Final readiness checklist

  • ComfyUI launches and opens in the browser
  • At least one model loads in the "Load Checkpoint" node
  • You've generated at least one image successfully
  • ComfyUI Manager is installed (the Manager button appears)
  • At least one ControlNet model is downloaded
  • You know how to relaunch ComfyUI tomorrow without help
→ Ready for Workshop Day 1

If you've ticked every box, you're set. See you in the first session. If anything broke, the contact page has WhatsApp — fastest channel for setup problems.

Next: read the Prompt Crafting Guide

Now that ComfyUI runs, the next skill is writing prompts that actually produce architecture rather than generic AI mush. That's a separate, longer guide.