Gigabyte R293-S90 2U 9-GPU Server (HGX H100 SXM5 / NVLink)

The Gigabyte R293-S90 is a 2U rackmount server purpose-built for GPU-accelerated AI, machine learning, and high-performance computing workloads. It accommodates up to 9 NVIDIA H100 SXM5 GPUs with full NVLink interconnect support, enabling low-latency peer-to-peer communication and unified memory semantics across the entire GPU cluster within a single node.

The system is architected around dual-socket NVIDIA GH200 Grace Hopper Superchip compatibility and supports AMD EPYC Genoa/Bergamo processors in baseline configurations. With 24 high-bandwidth NVLink connections per GPU and support for 36 GPU-to-GPU fabric lanes, the R293-S90 delivers exceptional throughput for large-scale transformer training, distributed inference, and multi-instance GPU computing scenarios.

Designed for enterprise data centers requiring dense GPU compute in constrained rack space, the R293-S90 integrates comprehensive thermal management, redundant power supplies, and enterprise-grade system monitoring to support continuous operation in mission-critical sovereign AI and cloud infrastructure deployments.

Ideal for

• Large language model (LLM) training and fine-tuning with distributed data parallelism and tensor parallelism
• Real-time generative AI inference serving with high throughput and low latency requirements
• Financial modeling and Monte Carlo simulations requiring massive parallel floating-point computation
• Scientific computing and computational fluid dynamics (CFD) for engineering design optimization
• Video processing, 3D rendering, and digital content creation pipelines
• Multi-tenant GPU virtualization and containerized ML workloads across enterprise clusters

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.