NVIDIA GB200 NVL4 AI Data Center
High-density AI data center computing platform combining 4 NVIDIA Blackwell GPUs and 2 NVIDIA Grace CPUs with NVLink-C2C interconnect for scalable AI, HPC, and large-model inference workloads.
Accelerated AI Computing with NVIDIA GB200 NVL4
The NVIDIA GB200 NVL4 is a powerful, single-node AI supercomputing platform that combines four NVIDIA Blackwell GPUs with two NVIDIA Grace CPUs. Designed for enterprise AI, large language models, and high-performance inference, GB200 NVL4 delivers exceptional compute density and memory bandwidth in a compact, liquid-cooled server form factor.
Unified CPU–GPU Architecture for Demanding AI Workloads
Each GB200 Grace Blackwell Superchip integrates one NVIDIA Grace CPU tightly coupled with two NVIDIA Blackwell GPUs using fifth-generation NVLink-C2C. In the NVL4 configuration, four Blackwell GPUs are interconnected with high-bandwidth NVLink, enabling the system to operate as a single, unified accelerator for training and inference workloads.
High Efficiency with Direct Liquid Cooling
The GB200 NVL4 platform is engineered for modern data centers, featuring direct liquid cooling to support its high power envelope while maintaining optimal thermal efficiency. This design enables sustained peak performance, reduced energy consumption, and improved rack density for enterprise and hyperscale AI deployments.
Key Features
- Powered by 4 NVIDIA Blackwell GPUs and 2 NVIDIA Grace CPUs
- Fifth-generation NVIDIA NVLink with ultra-high GPU-to-GPU bandwidth
- Unified CPU–GPU memory architecture for accelerated AI workloads
- Optimized for large language models, inference, and HPC applications
- Direct liquid cooling for sustained performance and energy efficiency
- Supports NVIDIA AI Enterprise software platform
- Integrated support for NVIDIA Quantum-X InfiniBand and Spectrum-X Ethernet
- Compact single-node AI supercomputing design
| MFG Number | GB200-NVL4 |
|---|---|
| Condition | Item Condition : Brand New |
| Price | $300,000.00 |
| Show Product Will Be Available Soon on FE | No |
| Product Card Description | The NVIDIA GB200 NVL72 is a liquid-cooled, data-centre-class AI server delivering unprecedented performance for generative AI, large language models and HPC workloads. With a 72-GPU NVLink domain, the system acts as a unified accelerator delivering up to 1,440 PFLOPS FP4, 720 PFLOPS FP8, and supports massive 13.4 TB HBM3e memory and 130 TB/s high-speed interconnect. Designed for real-time trillion-parameter inference (30× H100) and large-scale training (4× H100), this rack-scale solution enables hyperscale AI infrastructure with extreme throughput and efficiency. |
| Order Processing Guidelines | Order Processing Guidelines:
Inquiry First – Please reach out to our team to discuss your requirements before placing an order. |
