This course introduces the AMD Ryzen™ AI x86 Embedded platform and the software workflow used to prepare, optimize, and run AI workloads across heterogeneous compute resources.
Participants explore the roles of the CPU, integrated GPU, unified memory, and XDNA™ NPU, then apply the Ryzen AI software stack through guided hands-on activities. The course connects model preparation with practical on-device deployment for generative AI and computer vision workloads.
The emphasis of this course is on:
Understanding the Ryzen AI Embedded architecture and the roles of the CPU, integrated GPU, unified memory, and XDNA NPU
Explaining the Ryzen AI software stack, ONNX Runtime workflow, and execution provider concepts
Quantizing and optimizing models using AMD Quark
Running a local large language model using FastFlowLM
Running a real-time YOLO-based vision workload on a Ryzen AI platform
Bringing together model preparation, runtime selection, deployment, and basic performance considerations
