Stable Diffusion optimized for AMD hardware

Stability and AMD have teamed up to provide ONNX-optimized versions of the Stable Diffusion models, optimizing performance on AMD Radeon™ GPUs and Ryzen™ AI APUs. The collaboration has produced several accelerated models that are now available on Hugging Face, signified by the "_amdgpu" suffix. The four available models, including Stable Diffusion 3.5 Large, Stable Diffusion 3.5 Large Turbo, Stable Diffusion XL 1.0, and Stable Diffusion XL Turbo, offer significant performance improvements, with up to 2.6x and 3.8x faster inference compared to their base PyTorch models on AMD hardware.