AMD introduces the first-in-industry Stable Diffusion 3.0 AI model generator compatible with XDNA 2 NPUs, specifically crafted for Ryzen AI laptops, allowing for local execution.
A new text-to-image generator, the Stable Diffusion 3.0 Medium AI model, has been introduced by AMD and Stability AI. This model is designed to generate customizable stock-quality visuals, and it is optimized for use on AMD Ryzen AI laptops equipped with specific hardware.
Minimum System Requirements
To run the Stable Diffusion 3.0 Medium AI model, a laptop must be equipped with an AMD Ryzen AI 300-series or Ryzen AI MAX+ processor featuring an XDNA 2 NPU with at least 50 Tera Operations Per Second (TOPS) capacity. The system should also have a minimum of 24 GB of RAM, as the model uses approximately 9 GB during image generation. To achieve optimized local image generation, BF16 (Block FP16) precision computation via the XDNA 2 NPU is required.
With this configuration, creators and designers can generate high-quality, customizable images at up to 2048×2048 resolution, without the need for a dedicated GPU, cloud access, or an internet connection.
| Component | Minimum Requirement | |-----------------------|----------------------------------| | CPU | AMD Ryzen AI 300-series or Ryzen AI MAX+ | | NPU | XDNA 2 NPU with ≥50 TOPS | | System RAM | 24 GB | | Precision Format | BF16 (Block FP16) | | Image Resolution | Generates 1024×1024 images, upscaled to 2048×2048 |
Advanced Prompting Features
The Stable Diffusion 3.0 Medium AI model supports advanced prompting features, allowing users to fine-tune image composition. This gives creators and designers the ability to create a wide variety of customizable images to suit their needs.
Usage and Installation
To use the Stable Diffusion 3.0 Medium AI model, users must install the latest AMD Adrenalin Edition drivers and the Amuse 3.1 Beta software from Tensorstack. In Amuse, users should switch to EZ Mode, move the slider to HQ, and enable the 'XDNA 2 Stable Diffusion Offload' option.
An example prompt for the model to draw a toucan is provided: "Close up, award-winning wildlife photography, vibrant and exotic face of a toucan against a black background, focusing on the colorful beak, vibrant color, best shot, 8k, photography, high res."
Licensing
Usage of the Stable Diffusion 3.0 Medium AI model is subject to the Stability AI Community License. The model is free for individuals and small businesses with less than $1 million in annual revenue. However, licensing terms may change in the future.
Limitations and Caveats
As Amuse is still in beta, its stability and performance may vary. Generated images can be upscaled to 2048×2048 resolution, resulting in 4MP outputs. The model is optimized for use with the AMD Adrenalin Edition drivers and the Amuse 3.1 Beta software from Tensorstack.
In summary, the Stable Diffusion 3.0 Medium AI model offers a unique solution for creators and designers seeking fast, high-quality image synthesis locally on Ryzen AI laptops without relying on cloud services.
This new technology, the Stable Diffusion 3.0 Medium AI model, leverages artificial-intelligence capabilities to generate high-quality, customizable images on AMD Ryzen AI laptops equipped with specific hardware. Advanced prompting features allow users to fine-tune image composition, harnessing the power of artificial-intelligence to create a variety of customizable images.
