AMD's updated Ryzen AI Halo developer platform is now available at Micro Center for $7,299.99. The Linux-based mini PC features the Ryzen AI Max+ PRO 495 processor and 192GB of unified memory.
AMD’s upgraded developer platform, the Ryzen AI Halo, has reached retail availability at Micro Center for $7,299.99. The mini PC is designed for local AI development and features the Ryzen AI Max+ PRO 495 processor. This new configuration replaces the previous model, which launched in June at $3,999.99, making the updated system $3,300 more expensive.
The Ryzen AI Max+ PRO 495 processor includes 16 Zen 5 cores and 32 threads, with a boost clock of up to 5.2 GHz. The system integrates a Radeon 8065S GPU with 40 RDNA 3.5 compute units and an XDNA 2 NPU that delivers up to 55 TOPS. AMD rates the platform at up to 61 TFLOPS of FP16 GPU compute performance with a 120W TDP.
A key upgrade in this release is the memory capacity. The system comes with 192GB of LPDDR5X-8533 unified memory, which provides 273 GB/s of bandwidth. This is an increase from the 128GB supported by the previous Ryzen AI Max+ 395 model. AMD states that users can allocate up to 160GB of this memory as graphics memory. This configuration allows the system to run AI models with more than 300 billion parameters at 4-bit quantization locally, without relying on external cloud servers.
The Ryzen AI Halo includes a 2TB self-encrypting M.2 SSD for storage. Connectivity options include 10GbE networking, Wi-Fi 7, Bluetooth 5.4, and HDMI 2.1b output. The platform supports ROCm and various development tools such as PyTorch, Ollama, llama.cpp, and LM Studio.
Micro Center lists the Linux version of the Ryzen AI Halo as available for in-store purchase, with stock shown at several US locations. The $7,299.99 price applies to this Linux configuration. AMD also supports Windows 11 on the developer platform. A Windows configuration featuring Microsoft’s Project Zenith development environment is expected to follow later.
The launch of the Ryzen AI Halo with 192GB of memory provides developers with a dedicated workstation for running large AI models locally. By increasing the unified memory capacity, AMD addresses bandwidth bottlenecks associated with loading large neural networks. The system targets software developers building and testing machine learning architectures who require significant local processing power and memory without cloud dependency.
Sources
This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.
The day's checked news, by email
One short email a day with the top stories. Free, and you can unsubscribe in one click.
Or follow for the latest news:
TechTrending
Apple will hold a special event on October 13, 2026, titled 'Welcome home,' focusing on smart home hardware including the rumored HomeView hub and updated HomePod mini.
TechSponsored
Hutton Aerographics lists drone services in Kamloops, the BC Interior, and the Lower Mainland, including real estate photography, construction monitoring, and surveying.
