AMD Strix Halo Ryzen AI MAX APUs debut on mini-ITX boards with 128 GB LPDDR5X

Sixunited launches a compact mini-ITX board featuring AMD´s top-tier Ryzen AI MAX ´Strix Halo´ APUs, unified memory, and advanced Artificial Intelligence capabilities.

Chinese original design manufacturer Sixunited has unveiled the STHT1 Thin Mini-ITX board, delivering AMD´s highly anticipated Strix Halo ´Ryzen AI MAX´ APUs to desktop enthusiasts. Previously exclusive to premium laptops and compact PCs, these advanced Zen 5 system-on-chips offer up to 16 CPU cores, a Radeon 8060S graphics unit with up to 40 compute units built on the RDNA 3.5 architecture, and a dedicated integrated engine for Artificial Intelligence acceleration. The STHT1 board incorporates a non-socketed layout, enabling buyers to select from the entire Ryzen AI MAX lineup, including the entry-level six-core, 16-thread 380 SKU up to the flagship 16-core, 32-thread 395+ model. Each configuration supports a configurable thermal design power between 45 and 120 watts for versatile deployment scenarios.

Maximizing the performance envelope, Sixunited has equipped the STHT1 with a substantial 128 GB of LPDDR5X-8000 memory, soldered onboard to serve as a unified high-speed memory pool. This arrangement feeds the CPU, GPU, and AI engine simultaneously, crucial for bandwidth-intensive tasks such as large language model inference, advanced real-time rendering, and other demanding applications—all within a minimal desktop footprint. The board´s mobile-on-desktop (MoDT) approach uniquely addresses the need for both compactness and high computational throughput, a rare combination in enthusiast-grade hardware.

Connectivity and expandability are prominent features on the STHT1. The board supports two M.2 2280 PCIe 4.0 x4 slots for fast NVMe storage and an extra M.2 2230 slot for optional Wi-Fi modules. The rear panel comes equipped with an HDMI 2.1 port, and users can opt for either DisplayPort, VGA, or COM output depending on their needs. There are multiple USB options—including USB 3.2 Gen 2 and Gen 1 Type-A ports—alongside Gigabit Ethernet and audio jacks for comprehensive I/O flexibility. The power delivery system is robust, featuring a 16-phase voltage regulator for the SoC and additional phases for subsystem stability. With two fan headers, both a standard four-pin ATX connector and a 19-volt DC-in option, and everything contained in a compact 170×170 mm PCB, the STHT1 provides a versatile and high-performance platform for advanced desktop builds prioritizing Artificial Intelligence and graphics workloads.

65

Impact Score

Used Optane memory runs trillion-parameter model on one GPU

A workstation built with second-hand Intel Optane persistent memory modules was used to run Kimi K2.5 locally with a single GPU. The setup highlights renewed interest in a memory tier between DRAM and SSDs for large language model inference.

Colorado narrows its Artificial Intelligence law before rollout

Colorado is poised to significantly scale back its original Artificial Intelligence governance framework before it takes effect. The revised bill shifts the focus from broad upfront assessments to disclosure, recordkeeping, and post-deployment compliance.

UC Berkeley law tightens Artificial Intelligence rules without banning it

UC Berkeley Law is adopting a stricter Artificial Intelligence policy that limits student use in core legal work while preserving space for specialized courses and instructor discretion. Faculty behind the change say the aim is to protect foundational lawyering skills as generative tools become more capable and more common in practice.

EU eases parts of its Artificial Intelligence Act

The EU has agreed targeted changes to its landmark Artificial Intelligence Act, delaying some deadlines, narrowing parts of the high-risk category, and cutting overlapping compliance requirements. The package also adds a ban on tools that generate non-consensual sexually explicit images.

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.