AMD unveils Ryzen artificial intelligence Halo developer box at CES 2026

AMD is positioning its new Ryzen artificial intelligence Halo box as a compact desktop and full artificial intelligence development platform aimed at consumer applications, drawing a comparison to NVIDIA's DGX Spark. The system combines Strix Halo silicon with a custom cooling design and unified memory to attract developers targeting Windows and Linux.

AMD used the 2026 International CES to introduce its Ryzen artificial intelligence Halo box, a first-party compact desktop built around AMD Ryzen artificial intelligence Max+ ‘Strix Halo’ processors. The company is not presenting the system as only a consumer desktop, but as a fully fledged artificial intelligence development platform that AMD positions against NVIDIA’s DGX Spark. While AMD notes that the Ryzen artificial intelligence Halo box does not directly match the DGX Spark in compute power, the focus is on giving developers a local environment for building and testing consumer artificial intelligence applications.

Inside, the Ryzen artificial intelligence Halo box integrates a full-featured Ryzen artificial intelligence Max 395+ with a 16-core/32-thread ‘Zen 5’ CPU, a large iGPU based on the RDNA 3.5 graphics architecture with 40 compute units, and a Microsoft Copilot+ ready NPU that delivers 50 TOPS. These components are paired with up to 128 GB of unified LPDDR5X memory that is shared across the CPU, GPU, and NPU to simplify data movement for artificial intelligence workloads. The platform supports both Windows 11 and Linux, and AMD includes pre-loaded artificial intelligence models that are specifically optimized to take advantage of the hardware configuration inside the box.

To keep the Strix Halo system on chip operating efficiently in the compact chassis, AMD designed a custom cooling solution that uses a baseplate, a network of direct-touch flat heatpipes, an aluminium channel heatsink, and two lateral airflow blowers. This thermal design is intended to sustain the performance of the CPU, iGPU, and NPU under prolonged artificial intelligence development and testing workloads. AMD states that the Ryzen artificial intelligence Halo developer platform is scheduled to be available from Q2 2026, and the company indicates that pricing details will be announced closer to the launch window.

52

Impact Score

Adaptive training method boosts reasoning large language model efficiency

Researchers have developed an adaptive training system that uses idle processors to train a smaller helper model on the fly, doubling reasoning large language model training speed without sacrificing accuracy. The method aims to cut costs and energy use for advanced applications such as financial forecasting and power grid risk detection.

How to run MiniMax M2.5 locally with Unsloth GGUF

MiniMax-M2.5 is a new open large language model optimized for coding, tool use, search, and office tasks, and Unsloth provides quantized GGUF builds and usage recipes for running it locally. The guide focuses on memory requirements, recommended decoding parameters, and deployment via llama.cpp and llama-server with an OpenAI-compatible interface.

Y Combinator backs new wave of computer vision startups in 2026

Y Combinator’s 2026 computer vision cohort spans infrastructure, developer tools, and industry-specific applications from retail security to aquaculture and healthcare. Startups are increasingly pairing computer vision with large vision language models and foundation models to tackle real-time video, automation, and domain-specific analysis.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.