ASRock Industrial announced a strategic collaboration with Axelera AI to integrate Metis AIPU cards into its edge AI platforms, delivering high-performance, power-efficient AI inference solutions. The integration adds up to 214 TOPS of inference via M.2 or PCIe acceleration cards, significantly lowering power consumption and total cost of ownership compared to GPU-based architectures. Designed for LLM, machine vision, robotics, and smart city applications, the solution enables organizations to deploy advanced AI workloads at the edge with minimal footprint.

Validated configurations span expandable Edge AIoT platforms including the iEPF-11000S, iEPF-10000S, iEPF-9500S, iEPF-9040S, and iEPF-9040VS series with PCIe cards, plus compact systems like the AI BOX-A395, NUC BOX-358H, NUC BOX-255H, and NUC BOX-155H with M.2 cards. These support parallel AI workloads for real-time analysis across industrial and enterprise environments.

"Organizations are looking for practical ways to scale AI beyond proof-of-concept projects," said James Lee, Chairman of ASRock Industrial. Fabrizio Del Maffeo, CEO of Axelera AI, added that the collaboration provides a practical way to run demanding AI workloads within system constraints.

ASRock Industrial continues expanding its AI Pathfinder tool with validated hardware recommendations to simplify platform selection and accelerate deployment decisions for diverse AI workloads.