Difference between revisions of "Edge AI SDK/AI Framework/AMD"
| Line 72: | Line 72: | ||
[[File:image_1756953598517.png|none|900x600px|image_1756953598517.png]] | [[File:image_1756953598517.png|none|900x600px|image_1756953598517.png]] | ||
| − | + | | |
[[File:image_1757555234819.png|none|900x600px|image_1757555234819.png]] | [[File:image_1757555234819.png|none|900x600px|image_1757555234819.png]] | ||
| − | + | | |
[[File:image_1757555309652.png|none|900x600px|image_1757555309652.png]] | [[File:image_1757555309652.png|none|900x600px|image_1757555309652.png]] | ||
Revision as of 09:33, 18 September 2025
Contents
RyzenAI SDK
AMD Ryzen™ AI software includes the tools and runtime libraries for optimizing and deploying AI inference on AMD Ryzen AI powered PCs1. Ryzen AI software enables applications to run on the neural processing unit (NPU) built in the AMD XDNA™ architecture, the first dedicated AI processing silicon on a Windows x86 processor2, and supports an integrated GPU (iGPU).
Ryzen™ AI software enables developers to efficiently port their pre-trained PyTorch or TensorFlow models to run on the integrated GPU (iGPU) or Neural Processing Unit (NPU) available in select Ryzen AI-powered laptops.
Applications are deployed using ONNX Runtime, giving the user easy inferencing across the different hardware.
AMD Ryzen™ AI Software includes the tools and runtime libraries for optimizing and deploying AI inference on AMD Ryzen™ AI powered PCs. Ryzen AI software enables applications to run on the neural processing unit (NPU) built in the AMD XDNA™ architecture, as well as on the integrated GPU. This allows developers to build and deploy models trained in PyTorch or TensorFlow and run them directly on laptops powered by Ryzen AI using ONNX Runtime and the Vitis™ AI Execution Provider (EP).
Addressing Various AI Workloads -Large Language Models -Image and Video Generation -Recommendation -Computer Vision
Application
Vision Application
| Application | Model | AIMB-2210 FPS (video file) |
| Object Detection | yolov8-onnx-FP32 (CPU) | 25 |
| Object Detection | yolov8-onnx-FP32 (iGPU) | 35 |
| Object Detection | yolovX-onnx-int (NPU) | 65 |
| Face Detection | yolov8-onnx-FP32 (CPU) | 35 |
| Face Detection | yolov8-onnx-FP32 (iGPU) | 45 |
| Face Detection | RetinaFace-onnx-int (NPU) | 65 |
To Run NPU (YolovX)
Benchmark (Vision AI)
You can refer the linkto test the performance with the benchmark tool
mobilenet-ssd V2
System Monitoring
Lemonade (Generate AI)
Download Models from Lemonade
1. On Lemonade Model Management - Hot Models/Suggested Models, select a model and download it.
- Lemonade Model Management : http://localhost:23953/#model-management
- Support models on Ryzen AI 8000: GGUF.
2. Next, go to GenAI-Chatbot, where you can see the downloaded models.
- GenAI-Chatbot : http://localhost:23950/
3. Next, you can start chatting with the downloaded models.
Delete model
1. On Lemonade Model Management - Installed Models, select a model and delete it.
- Lemonade Model Management : http://localhost:23953/#model-management
