Get Started with AI on AMD Platforms

AMD is uniquely position to run AI models in the breath of its technology.

In this page, we will focus on AI inference on the different AMD technologies. We will show you how to run pre-trained models from popular model hubs such as HuggingFace, PyTorch Hub, etc. In the getting started we will mostly focus on the model functionality and how to run it on AMD technologies, model architecture and deeper understanding of the model internals will be covered in a different section.

PyTorch is our framework of choice for most of these resources due to its huge popularity, however, you will be able to find similar examples in any of the other popular AI frameworks.

Requisites

CPUs: AMD EPYC™ Processors AMD Ryzen™ Processors AMD Ryzen™ AI Processors

Or

GPUs: AMD Instinct™ Accelerators AMD Radeon™ RX Graphics Cards AMD Radeon™ PRO Graphics Cards

• Setup environment

Run Pre-trained Models

We provide a number of notebooks where you can run pre-trained models.

At the top of the notebook we will provide a tag of the technologies where the notebook can be run on using the following colors.

Direct support Can run on systems with enough memory

Check out:

• Using Hugging Face pre-trained models
  • MNIST Classification with an MLP Hugging Face Model
  • Semantic Segmentation
  • Image Classification using Yolov10
  • Fine-tuned LAnguage Net Text-To-Text Transfer Transformer
  • Google Enhanced Multimodal Machine Learning 2
  • OpenAI Whisper - Speech Recognition
  • Phi-3-vision Instruct Open Multimodal Model
  • Phi-3 Instruct Open Model
• Using PyTorch Hub Models
  • LeNet
  • Residual Network - ResNet50
• Generative AI on AMD platforms

More Resources

List of extra resources listed in no particular order.

• Microsoft Resnet-18 from Hugging Face

• Image classification using Vision Transformer with AMD GPUs

• Efficient image generation with Stable Diffusion models and ONNX Runtime using AMD GPUs

• Segment Anything with AMD GPUs

• text-generation-webui

• AUTOMATIC1111 stable-diffusion-webui

• How to run a Large Language Model (LLM) on your AMD Ryzen™ AI PC or Radeon™ Graphics Card

• Experience Meta Llama 3 with AMD Ryzen™ AI and Radeon™ 7000 Series Graphics

• Automatic1111 Stable Diffusion WebUI with DirectML Extension on AMD GPUs

• Developer Blog: Build a Chatbot with Ryzen™ AI Processors

• DBRX Instruct on AMD GPUs

• Deep Learning Recommendation Models on AMD GPUs

• LM Studio

• Llamafile

• Ollama

---

Copyright (C) 2025 Advanced Micro Devices, Inc. All rights reserved.

SPDX-License-Identifier: MIT