AMD Unveils a New AI Tool for Easy Image Generation on Consumer PCs

AMD has unveiled a new AI tool that helps PC users generate high-quality images on consumer devices running the company’s hardware.

The new Amuse 2.0, released in beta, is designed to provide an easy-to-install package to enable consumers to generate their own images.

Amuse can generate images from scratch from text inputs or existing images. It can also create images from rough sketches, similar to Nvidia’s GauGan tool.

Users can also create their own AI filters, changing the look and feel of existing images using generative AI.

Credit: AMD

Amuse generates images by leveraging Stable Diffusion XL, the powerful open source image generation model from Stability to enable consumers to easily and quickly create custom images.

The simplified user interface means users don’t have to set up configurations manually. Simply download and install Amuse. Once installed, users have to install Stable Diffusion XL, which can be done in a single click.

The program automatically detects the hardware fitted to a user’s PC and auto configures it to run on their device.

“Amuse 2.0.0 is an easy-to-install program, that has no dependencies, a simplified user interface called Ez Mode and will quickly take care of all the configuration required to jump-start your AI image generation journey, “according to an AMD blog post.

Related:AMD New Chips Take on Nvidia, Power AI PC Workloads

It does however require some higher-end graphics units to run. AMD recommends users have a Ryzen AI 300 series processor with at least 24GB of RAM or an 8040 series processor with at least 32GB of RAM

Not every consumer PC has such higher-end graphics cards, however. The application can still run on lesser hardware, just not very well.

The below image was created on a PC that featured an AMD Ryzen 5 3600 processor and an Nvidia GeForce GTX 1660 Super graphics card.

While it worked well, the processing time was far beyond what other image generation systems are capable of, taking upward of 20 seconds to create an image, whereas using Stable Diffusion XL takes around eight seconds.

Credit: AI Business via Amuse 2.0

For first-time image generator users looking to use Amuse, AMD recommends using the “Balanced” setting to provide an ample mix of performance and quality in outputs.

Like most image generation platforms, results vary, and can return bizarre images, such as the below’s attempt to “turn this dog into a distinguished gentleman wearing a top hat.”

Credit: AI Business via Amuse 2.0

Amuse 2.0 is of course just a beta at present, with AMD describing it as “experimental software.”

“It is being provided on an ‘as is’ basis,” according to a company blog post. “As such, there may be bugs and the user experience may be unstable, or cause unintended behavior.”

Related:AI Image-Generation Models and Tools: The Ultimate List