Deploy an Imaging AMD Solution Blueprint on AMD Radeon™ GPUs

2026-07-22 · Source: AMD ROCm Blogs · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure · Depth: Intermediate, long

Summary

AMD introduces technical preview support for AMD Radeon™ AI PRO R9700 and AMD Radeon™ PRO W7900 GPUs within its enterprise AI reference stack, released on July 22, 2026. This stack includes AMD Inference Microservices (AIMs), Solution Blueprints, AI Workbench, Resource Manager, and the AMD GPU Operator. These components enable users to serve LLMs, deploy reference applications like RAG chatbots, and simplify GPU management in Kubernetes clusters. The article demonstrates deploying the MRI Analysis Tool Solution Blueprint on an AMD Radeon AI PRO R9700S GPU in Kubernetes. This tool performs image preprocessing, tissue segmentation, anomaly detection, and LLM-assisted reporting using the Qwen3 VL 8B Instruct model, accessible via a Gradio web interface. Customization options, such as switching to the google/gemma-3n-E4B-it AIM and adjusting ephemeral storage from 120Gi to 200Gi, are also detailed.

Key takeaway

For AI Engineers or MLOps teams deploying AI workloads on AMD hardware, this blueprint simplifies integrating LLMs and complex applications into Kubernetes. You can utilize AMD's enterprise AI stack, including AIMs and Solution Blueprints, to serve models like Qwen3 VL 8B Instruct or google/gemma-3n-E4B-it. Consider using the AMD GPU Operator to automate GPU management and scale your generative AI deployments efficiently on Radeon AI PRO R9700 or PRO W7900 GPUs.

Key insights

AMD's enterprise AI stack enables streamlined deployment and management of AI models and applications on Radeon GPUs within Kubernetes.

Principles

• Standardized microservices simplify AI model serving.
• Kubernetes operators automate GPU resource management.
• Solution Blueprints offer complete microservice examples.

Method

Deploy AMD Solution Blueprints by generating a Kubernetes manifest from an OCI-compliant Helm chart using `helm template`, then applying it with `kubectl apply -f -`. Customize via `--set` flags or override files.

In practice

• Serve LLMs using AMD Inference Microservices (AIMs).
• Deploy RAG chatbots from Solution Blueprints Catalog.
• Manage Radeon GPUs in Kubernetes with AMD GPU Operator.

Topics

• AMD Radeon GPUs
• Kubernetes Deployment
• AI Microservices
• Solution Blueprints
• LLM Inference
• GPU Operator

Code references

• amd-enterprise-ai/solution-blueprints

Best for: MLOps Engineer, AI Engineer, AI Architect

Related on AIssential

• Deploy and Customize AMD Solution Blueprints — AMD Solution Blueprints enable efficient, customizable deployment of AI microservices on AMD Enterprise AI Suite clusters.
• Day 15 of 100 Days of GenAI for DevOps: GenAI on Kubernetes — Kubernetes requires specific components like Device Plugins to enable GPU support for AI workloads.
• Solution Blueprints: Accelerating AI Deployment with AMD Enterprise AI — AMD Solution Blueprints streamline AI deployment on Instinct GPUs via modular Helm charts and pre-integrated AIMs.
• GPU Clouds, Aggregators, and the New Economics of AI Compute — GPU cloud providers offer diverse services and pricing, necessitating careful evaluation based on security, managed services, and workload portability.
• AMD: Advancing AI Data Centres, Networks and Robots — AMD's strategy focuses on an open ecosystem across hardware and software to accelerate AI development and deployment from data centers to edge robotics.
• Maximizing GPU Utilization: Heterogeneous Pipelines with Ray and Kubernetes — Ray optimizes heterogeneous, distributed AI workloads by managing diverse compute resources and handling failures across complex, multi-layered stacks.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AMD ROCm Blogs.