
The series brings enterprise‑grade AI and graphics power to edge and embedded environments, enabling faster, locally processed workloads for industrial automation and vision applications.
Edge computing is reshaping how manufacturers and system integrators handle data, and the demand for powerful yet space‑efficient hardware has surged. ASRock’s NUC Ultra 300 Box series arrives at a moment when traditional mini PCs struggle to meet AI‑driven workloads. By leveraging Intel’s third‑generation Core Ultra silicon, the new boxes combine a heterogeneous core layout with a dedicated NPU, delivering up to 180 TOPS of AI throughput that rivals many larger form‑factor servers while staying under 50 mm tall.
The technical specifications reinforce the series’ ambition. The Core Ultra X7 358H variant packs a 4‑8‑4 mix of performance, efficient, and low‑power cores, paired with an Arc B390 graphics module featuring twelve Xe³ cores for both visual rendering and GPGPU tasks. Memory bandwidth jumps to DDR5‑7200, supporting a maximum of 128 GB, and two M.2 slots enable high‑speed NVMe storage alongside an optional Wi‑Fi 7 module. Connectivity options span USB4/Thunderbolt 4, USB 3.2 Gen 2×2, and dual 2.5 Gbps Ethernet on the larger chassis, ensuring the devices can serve as central nodes in rugged industrial networks.
For the industrial market, these capabilities translate into tangible benefits. Edge AI inference for quality inspection, predictive maintenance, or autonomous robotics can now run locally with sub‑millisecond latency, reducing reliance on cloud bandwidth and enhancing data security. The modular barebone approach lets OEMs tailor RAM, storage, and wireless components to specific project budgets, while the compact footprint eases integration into confined enclosures. As competitors roll out similar AI‑enabled edge boxes, ASRock’s early entry with a robust processor‑GPU‑NPU trio positions it as a compelling choice for enterprises seeking to future‑proof their embedded deployments.
Comments
Want to join the conversation?
Loading comments...