AI Inference Server
Data center-grade servers for large-scale AI inference.
Overview
The AI Inference Server is a high-density, GPU-optimized server platform designed for large-scale AI inference and model serving in data center environments. Supporting up to 8 double-width GPU accelerators and 4TB of system memory, it delivers unmatched throughput for LLM inference and enterprise AI workloads.
Processor
Dual Intel Xeon 6 / AMD EPYC, up to 288 cores
Memory
32x DDR5 DIMM slots, up to 4TB ECC
Storage
12x NVMe U.2 hot-swap bays, 2x M.2 boot drives
GPU / AI Accelerator
Up to 8x GPU Accelerators (supports multiple vendors)