View all AI news articles

Unleashing Creativity with Stable Diffusion Generative AI on NVIDIA RTX 3090

May 17, 2024


Generative AI has been a hot topic in the tech world, offering a plethora of opportunities for artists, designers, and developers. One such fascinating application is Stable Diffusion, a generative AI model that allows you to create stunning images. In this article, we'll walk you through the journey of setting up Stable Diffusion on an NVIDIA RTX 3090 GPU, right from building the machine to generating your first piece of AI art.

Building the Machine

The first step in this journey is to build a machine capable of running Stable Diffusion. For our example, we used an NVIDIA RTX 3090 GPU, which offers excellent performance for generative AI tasks. It took about a day to assemble the hardware and configure the machine, ensuring that it was ready for the software installation.

Software Installation

The software setup involves installing Python, Git, and other prerequisites. The Medium article by Fanis Spyrou provides a step-by-step guide for installing Stable Diffusion on a fresh Windows 10 build. The article covers everything from installing Python and Git to setting up the Stable Diffusion web UI.

Steps to Follow:

  1. Install Prerequisites: Download and install Python 3.10.6 and Git. Make sure to add Python to PATH during the installation process.
  2. Clone the Repository: Use PowerShell to clone the Stable Diffusion web UI repository from GitHub.
  3. Download Models: Download one or more Stable Diffusion model checkpoints and place them in the appropriate directory.
  4. Start the Web UI: Run the batch file to start the web UI and wait for the model weights to load.

Experimenting with Prompts

Once the software is set up, the next step is to experiment with prompts to generate images. This process took a couple of hours, as it involves tweaking the settings and observing the output. The Stable Diffusion web UI offers a user-friendly interface to make this process easier.

Generating the Image

After experimenting with various prompts, it took just 48 seconds to generate a particular image using the NVIDIA RTX 3090 GPU. The speed and efficiency of the GPU play a significant role in the quick generation of high-quality images.

Alternative Options

If you don't have access to a powerful GPU, you can rent one by the hour from platforms like


What is Stable Diffusion?

Stable Diffusion is a generative AI model that allows you to create images based on textual prompts.

How long does it take to set up Stable Diffusion?

It takes about a day to build and configure the machine, a couple of hours to experiment with prompts, and less than a minute to generate an image.

Can I run Stable Diffusion on a less powerful GPU?

Yes, you can rent a GPU by the hour from platforms like if you don't have a powerful GPU.


Stable Diffusion offers a fascinating avenue for creative expression, made even more accessible and efficient with powerful GPUs like the NVIDIA RTX 3090. Whether you're an artist looking to explore new mediums or a tech enthusiast interested in generative AI, Stable Diffusion provides an easy and efficient way to dive into the world of AI-generated art.

Recent articles

View all articles