 |
| AMD counters Nvidia's DGX Spark with a mini-PC powered by Ryzen Strix Halo. |
At CES 2026, the battle for the AI developer’s desk is heating up. AMD has just pulled back the curtain on the Ryzen AI Halo, a mini-PC powerhouse designed as a direct competitor to Nvidia’s popular DGX Spark AI workstation. This move signals AMD's full-throated entry into the lucrative market for compact, high-performance AI development systems.
While Nvidia’s solution famously pairs an ARM processor with a potent graphics chip for 1,000 TOPS of AI performance, AMD is taking a different architectural path. The Ryzen AI Halo is built around the formidable Ryzen AI Max+ 395 chip, a system-on-a-chip (SoC) that promises to blend raw computing power with exceptional AI capabilities in a sleek, quiet package.
The Heart of the Machine: Ryzen AI Max+ 395 Unpacked
At its core, the Ryzen AI Max+ 395 is a technical marvel. It combines 16 high-performance CPU cores capable of boost clocks up to 5.1 GHz with an integrated Radeon 8060S graphics unit featuring 40 compute units. This integration isn’t just about saving space; it’s about efficiency and shared resources.
Surprisingly for an AI workstation, AMD is also touting serious gaming chops. The company claims the SoC delivers enough performance to smoothly run demanding titles like Assassin’s Creed Shadows, The Elder Scrolls IV: Oblivion Remastered, and Cyberpunk 2077 at 1080p resolution with high to maximum detail settings. This versatility makes the Halo a potential all-in-one solution for developers who also appreciate premium gaming.
Built for the AI Frontier: Memory is King
However, gaming is a side benefit. The true target audience is AI researchers and developers. Here, AMD’s architecture makes a strategic play: both the CPU and the integrated GPU can directly access a shared pool of up to 128 GB of ultra-fast RAM.
This unified memory approach is a game-changer for local AI model work. AMD states this should enable developers to run models with up to 200 billion parameters locally on the Ryzen AI Halo. Eliminating the need for constant cloud computing or complex multi-GPU setups for large models could significantly streamline workflows and lower iteration times. This strategic focus on memory bandwidth and capacity is central to AMD’s expanding vision for AI leadership across its product stack.
Design Meets Demanding Thermals
Packing this much performance into a mini-PC requires serious cooling. The Ryzen AI Max+ 395, part of the "Strix Halo" family, can consume over 180 watts under load. AMD’s solution employs a robust cooling system featuring two fans and a large copper heatsink. Discreet ventilation openings are thoughtfully integrated across most of the anodized aluminum housing, which is subtly branded with two AMD logos. A small, customizable LED strip adds a personal touch without being overpowering, maintaining a professional aesthetic suitable for any office or lab.
The system will support both Windows 11 and Linux, catering to the preferred environments of most developers and researchers.
Price, Availability, and the Competitive Landscape
The AMD Ryzen AI Halo is expected to launch in the second quarter of 2026 (April-June). AMD has not yet announced official pricing, leaving the industry to speculate.
For a potential benchmark, we can look at current systems using the same core silicon. The GMKtec EVO-X2, also based on the Ryzen AI Max+ 395, is currently listed for $1,840 with 64 GB of RAM on retailers like Amazon. Given the Halo’s premium cooling, design, and potential for higher RAM configurations, a price point above this, yet competitive with Nvidia’s DGX Spark, is anticipated.
With the Ryzen AI Halo, AMD isn’t just offering an alternative; it’s offering a compelling argument for a unified memory architecture in local AI development. As Q2 2026 approaches, all eyes will be on final pricing and performance benchmarks to see if this halo product can truly dim the spark of its established rival.