Ryzen AI Halo mini PC, prepared by AMD for local artificial intelligence development, has come to the fore again with a June 2026 release schedule. The compact developer platform that the company demonstrated at CES 2026 is in a class that will directly compete with NVIDIA DGX Spark with its Ryzen AI Max 395 processor, up to 128 GB combined memory, ROCm support and Windows / Linux compatibility. AMD’s official calendar indicates the second quarter of 2026 for the product.
The June claim also coincides with the last month of this window. AMD Ryzen AI Halo mini PC is coming for local artificial intelligenceAMD Ryzen AI Halo is a small form factor artificial intelligence platform prepared for developers rather than a classic mini PC. AMD describes the product as a one-stop reference solution for native AI development. The platform includes optimized applications and AI models, Windows and Linux support, full AMD ROCm support, day one support for leading AI models, up to 128 GB of unified memory, and the Ryzen AI Max 395 processor.
At the heart of Ryzen AI Halo is the AMD Ryzen AI Max 395. This processor comes with 16 Zen 5 cores and 32 threads. According to AMD’s official technical data, the base frequency of the processor is 3.0 GHz and the maximum boost frequency is 5.1 GHz. The chip includes 16 MB L2 cache, 64 MB L3 cache, 55 W default TDP and a configurable TDP range of 45-120 W. TSMC 4 nm production technology is used on the CPU and I/O side.
Radeon 8060S is on the graphics side. Ryzen AI Max 395 comes with 40 AMD RDNA 3.5 compute units. On the artificial intelligence side, there is XDNA 2 NPU and AMD talks about 50 peak AI TOPS level for Ryzen AI Max 395. This structure offers a high-performance APU layout where the CPU, GPU and NPU are combined on the same platform. The most critical aspect of Ryzen AI Halo is the unified memory architecture. AMD states that the platform has up to 128 GB of unified memory and that large productive artificial intelligence models can be run locally.
AMD’s official statement also includes information that Ryzen AI Halo can run models with up to 200 billion parameters locally. This distinction moves the product from the ordinary mini PC segment to the artificial intelligence development station class that fits on the desktop. On the software side, the most important pillar of the platform is ROCm support. AMD states that it offers Ryzen AI Halo with the latest ROCm software and a ready experience for AI developer workflows.
Full AMD ROCm support, optimized applications and models, Windows/Linux compatibility and first-day model support are clearly listed on the product page. Developer tools such as LM Studio, ComfyUI and VS Code stand out in the shared information. Optimizations for models such as GPT-OSS, FLUX.2 and SDXL are also mentioned. Therefore, Ryzen AI Halo is gaining ground not only with its hardware power, but also with its ready-made software environment for native model running and generative artificial intelligence development process.
AMD has officially given the second quarter 2026 calendar for Ryzen AI Halo. New information shows that the product may arrive in June. This date is consistent with AMD’s previously announced Q2 2026 window. However, there is no official statement on the price side yet. AMD stated that price and commercial availability details will be shared closer to the launch. The most prominent product against Ryzen AI Halo is NVIDIA DGX Spark.
NVIDIA’s desktop artificial intelligence system comes with 128 GB unified system memory and is highlighted with the information that it can fine-tune models with up to 70 billion parameters. On the retail side, price information of $ 4,699 for DGX Spark has been brought to the agenda. Therefore, AMD’s official price will directly determine the position of Ryzen AI Halo in the market. It seems that AMD has established the main difference in this product through its open software ecosystem and the flexibility of the Ryzen AI Max platform in the PC world.
Windows and Linux support, developments on the ROCm side, unified memory layout and compact chassis create a separate product class for users doing local artificial intelligence development. Competition in this class is shaped not only by processing power, but also by the number of supported models, maturity of software tools, memory capacity, price and long-term driver support. It is clear that Ryzen AI Halo also has strong hardware on the gaming or standard desktop use side.
But the product’s bottom line focuses on native AI development, model running, generative AI workflows, and small form factor developer systems. 16-core Zen 5 processor, Radeon 8060S graphics unit, XDNA 2 NPU and up to 128 GB combined memory are among the main features that distinguish this device from classic mini PCs.
Comments
You can write your views about this story. Comments may be moderated according to site settings.