Supermicro’s latest 5U PCIe GPU servers are optimized to deliver the full power of AMD’s new Instinct™ MI350P PCIe GPUs, giving enterprises a fast, cost-effective way to scale generative and agentic AI workloads without redesigning their data centers.

AI Performance That Fits Your Racks

The new AMD Instinct MI350P PCIe cards are designed as drop-in accelerators for standard air-cooled servers. Supermicro’s AS-5126GS-TNRT and AS-5126GS-TNRT2 servers take full advantage of this flexibility, supporting up to 10 MI350P GPUs per server while staying within existing power and cooling envelopes.

Powered by dual AMD EPYC 9005 processors (up to 384 cores), these systems deliver exceptional performance and efficiency for AI inference, RAG pipelines, and agentic AI workloads — all within your current rack infrastructure.

Maximum Performance. Maximum ROI

By combining Supermicro’s high-density PCIe GPU platform with the MI350P’s industry-leading 144 GB HBM3e memory, 600W TDP, and support for low-precision MXFP4/MXFP6 formats, enterprises gain:

• Superior tokens-per-dollar economics and lower total cost of ownership

• Higher throughput per server, reducing footprint and licensing costs

• Faster time from evaluation to production

Built on an Open, Enterprise-Ready Ecosystem

Supermicro servers using AMD Instinct MI350P GPUs and the open-source ROCm software stack give you full vendor independence and zero licensing fees. The fully integrated AI software stack supports PyTorch, TensorFlow, vLLM, and more — making deployment and workload migration simple and cost-effective.

Server Highlights

See the Supermicro 5U PCIe GPU servers using AMD Instinct MI350P GPUs delivering enterprise AI performance firsthand — watch our new video below to experience the drop-in deployment, up to 10-GPU density, and high-throughput inference performance inside existing racks.

• AS-5126GS-TNRT: Dual-root PCIe architecture with direct 16-lane CPU-to-GPU connectivity for 8 GPUs

• AS-5126GS-TNRT2: PCIe-switched dual-root topology supporting up to 10 GPUs

• Up to 8 or 12 hot-swap NVMe/SATA drives (front-accessible) for flexible storage

• Designed for seamless integration into standard air-cooled racks

Deploy Enterprise AI Where You Are Today

With Supermicro and the new AMD Instinct MI350P PCIe GPUs, organizations can immediately scale AI capabilities inside their existing data centers — delivering more performance, better efficiency, and faster ROI.

For more information on Supermicro’s AMD Instinct MI350P solutions, visit: www.supermicro.com/aplus