SHARK team in collaboration with AMD demonstrated an ultra fast implementation of Stable Diffusion on the AMD RDNA3 GPUs running on Windows 11 at the AMD RNDA3 launch event

Stable Diffusion on AMD RDNA3 GPUs

SHARK is an open source cross platform (Windows, macOS and Linux) Machine Learning Distribution packaged with torch-mlir (for seamless PyTorch integration), with IREE (for efficient codegen, compilation and runtime) and’s tuning.

Stay “tuned” to experience the power of Generative AI and hundreds of other SOTA Machine Learning models in the SHARK tank on AMD’s RDNA3 GPUs when they become available.

Special thanks to the Google IREE team and AMD engineers for making this a reality. is hiring: Join us on Discord

Comments are closed.