ASUS Unveils XA P2N-E2 Server for Agentic AI Workloads with NVIDIA Vera CPU

ASUS is expanding its AI infrastructure portfolio with the introduction of the XA P2N-E2 server. This new 2U system is based on the NVIDIA MGX architecture and utilizes NVIDIA Vera CPUs, targeting agentic AI and reinforcement learning workloads. The product was unveiled at OCP Global Summit 2026.

The server focuses on the execution layer of AI applications, handling tasks such as code execution, tool use, sandboxing, data processing, and orchestration that go beyond the AI model itself. The XA P2N-E2 supports up to two dual-slot NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, combining high-performance CPU computing with GPU acceleration.

At the event, ASUS also showcased a broader range of solutions, including rack-scale AI systems. The ASUS AI POD, XA VR721-E3, integrates 36 NVIDIA Vera CPUs and 72 NVIDIA Rubin GPUs via the NVIDIA Vera Rubin NVL72 platform, featuring a fully liquid-cooled architecture for large-scale AI training and inference.

The company aims to provide scalable AI infrastructure, from individual servers to large environments, accommodating various compute densities, cooling architectures, and deployment requirements. This reflects ASUS's approach to building full-stack AI factories with the NVIDIA Vera Rubin platform.