Hi, I’m Jakob Stewart, a digital artist specializing in NSFW AI tools.
Today, I will teach you how to install and run Flux AI, a powerful new text-to-image model, especially if you're working with limited VRAM.
I'll break down the steps for installing Flux AI on platforms like ComfyUI and ForgeUI so you can easily generate amazing images, even on lower-end GPUs.
Click the links below to jump directly to the section.
What is Flux AI?
Flux AI is a cutting-edge text-to-image diffusion model developed by Black Forest Labs.
It outperforms major models like Stable Diffusion 3 Ultra and MidjourneyV6.0.
With 12 billion parameters, it’s powerful but can still run on lower-end machines, making it accessible to more users.
Flux comes in several versions:
Flux.1 Dev: Great for high-end GPUs with over 12GB VRAM.
Flux.1 Schnell: Optimized for lower-end GPUs, generating images in 1-4 steps.
Flux NF4: A 4-bit quantized version for low VRAM machines.
Installing Flux AI with Stability Matrix
Installing Flux AI is made simple by using Stability Matrix, a multi-platform package manager designed for Stable Diffusion web UI packages.
It provides a graphical user interface that makes it easier to install models without needing a command-line experience.
Check out my detailed guide on how to install Stability Matrix for step-by-step instructions to get started.
Step 1: Download Stability Matrix
Head over to the Stability Matrix download page and choose the installer that matches your operating system (Windows, Linux, or Mac).
Once downloaded, unzip the file and install Stability Matrix.
You can choose the portable version if you prefer to move the program folder to another drive while keeping all your data intact.
Step 2: Download the Flux AI Model
After installing Stability Matrix, you'll need to manually add the Flux AI model.
Use the following links to download the appropriate files:
VAE file for auto-encoding: Download the VAE here (Only if using Flux.1 Dev NF4)
⚠️ Once downloaded, place the Flux model file in the data/stable-diffusion folder and the VAE file in the data/vae folder within Stability Matrix.
Step 3: Start Stability Matrix and Install Web UI
Open Stability Matrix. When it launches for the first time, you'll be prompted to install packages.
Select the Stable Diffusion Web UI Forge package.
If the pop-up doesn't appear, you can always go to the Packages section and click Add Package to install the web UI with just one click.
Step 4: Select Flux AI in Web UI
Once the Stable Diffusion Web UI is installed, launch it and:
Check the Flux AI box at the top of the interface.
Select the Flux model and the corresponding VAE file from the list. (Only if using Flux.1 Dev NF4)
No further configuration is needed at this point, making it super simple to start generating images.
Step 5: Generate Your First Image
Now that Flux AI is set up, you can enter your text prompt in the designated area. If you're working with a low-VRAM GPU, make sure to reduce the image dimensions (e.g., 512 x 512) and keep the sampling steps low (1-4 for Schnell or 20-50 for Dev).
Click Generate and wait for the image to render. Depending on your GPU, it might take a few minutes, but Stability Matrix will handle all the heavy lifting for you!
Tips for Optimizing VRAM Usage:
Reduce Image Size: Smaller images, such as 512 x 512, consume less VRAM.
Lower Sampling Steps: Using fewer steps (1-4) can speed up the process and conserve VRAM.
Use the FP8 Version: The FP8 model is another way to maintain quality while saving resources.
I’ve tried these steps myself, and trust me, you’ll be amazed by what Flux AI can do. Give it a try and start creating today!
Want more tutorials like this? Subscribe to my blog for regular updates on NSFW AI tools and image generation tips!