Launching AMD AI Playbooks: Step-by-Step Guides for Building with AI Locally with AMD

Introducing AMD AI Playbooks: a hands-on learning destination for developers who want to build and real AI workflows locally on AMD hardware.

Each playbook walks through a complete, working setup from environment configuration to running models and building something useful on top. The focus is simple: pick a workflow, follow the steps, and get it running on your own machine.

From inference and automation to local generative AI workflows, Playbooks are designed to help developers move quickly from getting started to getting results.

What Are AMD AI Playbooks?

Playbooks are published as open guides on GitHub and cover end-to-end workflows, including:

• Environment setup and prerequisites
• Dependency and driver installation
• Running models locally on AMD hardware
• Building or extending a practical application

The Playbook Approach

Each playbook is built around a specific task and gets you to a working result quickly. Along the way, you’ll see how the pieces fit together without getting lost in theory.

The goal is to help you run something real, understand it, and build on it.

Who they are for?

AI Playbooks are written for developers who are comfortable with the basics:

• The command line
• Python environments
• Basic AI/ML concepts

You don’t need deep expertise in every tool used. Just enough to follow along and experiment.

First Available Playbooks

The initial set of AMD AI Playbooks covers real-world workflows developers are already exploring:

1. ComfyUI + Z Image Turbo - Generate images locally on AMD hardware
2. n8n + local LLMs - Build AI-powered automation workflows
3. VS Code + Qwen3-Coder - Run a local coding assistant on-device
4. LM Studio - Serve and interact with local LLMs
5. PyTorch + ROCm™ software - Run and accelerate LLM workloads, including tasks like summarization

More playbooks will continue to expand coverage across tools, models, and hardware.

Getting Help and Contributing Back

If you run into issues or want to connect with other developers building on AMD:

• Join the AMD Developer Community on Discord or Forums to connect with other developers, get technical guidance, and join community discussions.
• Report bugs, request features, or submit pull requests directly through the GitHub repository.

Feedback and contributions help shape future playbooks and improve the overall developer experience.

What's Next

On a regular basis, we’ll be publishing more playbooks as well as adding more AMD hardware coverage. If you’re building local AI apps, GPU-accelerated pipelines, or coding workflows on AMD Ryzen™ AI PCs, the Playbooks site is now live and ready for you.

Start with the AMD AI Playbooks that match your stack and run it end-to-end.

Join AMD AI Developer Program to access AMD developer cloud credits, expert support, exclusive training, and community.