Hi @rbg ,
The Gemma 3n 4B model isn't currently listed among the pre-optimized models for ONNX Runtime GenAI Flow. However, you can still deploy it on your AMD Ryzen AI 365 laptop by converting the model to the ONNX format and optimizing it using ONNX Runtime's Model Builder. This tool supports Gemma models and can help tailor them for your specific hardware.
Regarding compatibility, AMD Ryzen AI Software 1.4 supports the Ryzen AI 365 processor on Windows 11, provided your system meets the necessary prerequisites, including Visual Studio 2022, CMake 3.26 or higher, and Anaconda or Miniconda. Ensure your Windows build is 22621.3527 or newer.
While Gemma 3n 4B isn't officially pre-optimized for this setup, with the right tools and configurations, you can achieve efficient deployment on your system.
Best regards.