This repository provides demos, examples, and tutorials for optimizing and deploying AI inference on AMD Ryzen™ AI PCs. It targets developers looking to build and run AI applications efficiently, leveraging Ryzen™ AI's acceleration capabilities.

How It Works

The software enables AI acceleration on Ryzen™ AI hardware through ONNX Runtime, utilizing the Vitis-AI Execution Provider (EP). This approach allows for efficient execution of AI models, including LLMs and computer vision tasks, by mapping operations to the dedicated NPU and iGPU.

Quick Start & Requirements

• Install Git LFS from git-lfs.github.com.
• Run git lfs install to set up Git LFS.
• Clone the repository: git clone https://github.com/amd/RyzenAI-SW.git
• Pull LFS files: git lfs pull
• Follow individual READMEs within subdirectories for specific demo/example setup.
• Requires AMD Ryzen™ AI hardware.

Highlighted Details

• Includes examples for LLMs (e.g., GenAI API, BF16 models), text classification (DistilBERT), and real-time object detection (Yolov8).
• Demonstrates NPU-GPU pipelines and concurrent AI application execution with ONNX Runtime.
• Provides tutorials for fine-tuned ResNet models and ONNX benchmark utilities.
• Offers guidance on quantization and using the Ryzen AI CVML library.

Maintenance & Community

• References the "Ryzen™ AI Developer Guide" and "AMD AI Developer Forum" for further information.

Licensing & Compatibility

• License details are not explicitly stated in the provided README snippet. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

This repository is a subset of the full Ryzen™ AI Software release, and specific installation instructions are directed to the main software package. The README does not detail the specific license under which the software is distributed.