An 8-GPU PCIe server from Supermicro with NVIDIA RTX PRO 6000 Server Edition GPUs. 96 GB GDDR7 per GPU. Available with Intel Xeon 6900 or AMD EPYC 9005 processors.
Eight NVIDIA RTX PRO 6000 Server Edition GPUs with 96 GB GDDR7 per GPU and 768 GB total GPU memory. PCIe form factor with optional NVLink Bridge for GPU-to-GPU connectivity.
Choose Intel Xeon 6900 series (up to 128C/256T) or AMD EPYC 9005 series (up to 192C/384T). Up to 6 TB DDR5 system memory across 24 DIMM slots.
Supports up to 2 quad-width, 2 triple-width, 8 double-width, or 10 single-width GPUs. The 5U chassis accommodates multiple GPU form factors for different workload profiles.
Built on Supermicro's enterprise server platform with air cooling, Titanium-rated power supplies, and a global track record of reliability in production environments.
The NVIDIA RTX PRO 6000 Server Edition is a Blackwell-architecture PCIe GPU designed for data center, lab, and professional workstation environments. Unlike the HGX-based SXM platforms used in large-scale training clusters, the RTX PRO 6000 uses standard PCIe slots, making it compatible with a broader range of server chassis and deployment scenarios.
With 96 GB GDDR7 memory and Blackwell-generation compute performance, the RTX PRO 6000 is well suited for AI development, model fine-tuning, inference serving, 3D rendering, and professional visualization workloads that do not require the multi-GPU NVLink interconnect of HGX platforms.
Every cluster is designed around your workloads, your facility, and your operational model.
Systems are validated, burned in, and shipped with full deployment support.
Hands-on experience designing and deploying high-density GPU infrastructure.
From planning and procurement through deployment, operations, and scaling.
