NVIDIA GB200 Grace Blackwell Superchip combining 1 Grace CPU and 2 Blackwell GPUs over NVLink-C2C with 372 GB HBM3e and large coherent CPU-GPU memory for AI and HPC systems.
Fast Shipping
100k+ SKUs
8/5 Support
Certified
NVIDIA GB200 Grace Blackwell Superchip | GB200-SUPERCHIP
NVIDIA GB200 Grace Blackwell Superchip (GB200-SUPERCHIP)
NVIDIA GB200 Grace Blackwell Superchip combines one NVIDIA Grace CPU with two NVIDIA Blackwell GPUs using high-bandwidth NVLink-C2C. It is the compute building block behind platforms such as GB200 NVL72 and is designed for tightly coupled CPU-GPU workloads where memory bandwidth, coherent data movement, and multi-GPU scale-up are central to performance.
Availability / RFQ: This product is a current NVIDIA platform or component and is listed for RFQ. Pricing, allocation, quantity, configuration, support, and lead time can vary with current channel conditions. Contact Catalyst Data Solutions to discuss current status, compatibility, configuration requirements, and sourcing options.
End-user / export compliance notice: Advanced AI accelerators and systems may be subject to applicable U.S. export controls, end-user/end-use restrictions, destination requirements, and OEM/channel policies. Catalyst may request purchaser, final end-user, intended-use, and deployment-location information before a quote or fulfillment can be approved. An RFQ does not guarantee allocation or shipment.
Technical Specifications
| SKU | GB200-SUPERCHIP |
|---|---|
| Product Type | CPU-GPU Superchip |
| Configuration | 1x NVIDIA Grace CPU + 2x NVIDIA Blackwell GPUs |
| GPU Memory | 372 GB HBM3e |
| GPU Memory Bandwidth | 16 TB/s aggregate |
| CPU | 72-core Arm Neoverse V2 NVIDIA Grace CPU |
| CPU Memory | Up to 480 GB LPDDR5X |
| CPU Memory Bandwidth | Up to 512 GB/s |
| NVLink-C2C | Up to 900 GB/s CPU-GPU coherent interconnect |
| Availability Status | Available / RFQ |
Deployment and Compatibility Considerations
- GB200 is normally purchased as part of a validated OEM system, compute tray, or NVL architecture rather than as a generic socketed component.
- Confirm OEM board, thermal solution, power, firmware, management, and system-memory configuration.
- For cluster deployments, validate scale-out networking, storage, optics, cabling, and rack cooling.
Common Use Cases
- Grace Blackwell AI servers
- Large-model inference and training
- Accelerated data processing
- HPC and AI convergence
- GB200 NVL72 compute trays
Frequently Paired Infrastructure
- GB200 NVL72 and GB200 NVL4 platforms
- ConnectX networking
- BlueField-3 DPUs
- Spectrum-X800 or Quantum-X800
- Liquid-cooled OEM systems
Official NVIDIA Documentation
Nvidia GB200 Superchip Specifications
Frequently Asked Questions
What processors are combined in the GB200 Superchip?
GB200 combines one NVIDIA Grace CPU with two NVIDIA Blackwell GPUs over NVLink-C2C.
Is GB200 a conventional PCIe GPU card?
No. It is a tightly integrated CPU-GPU superchip used inside validated Grace Blackwell systems and rack-scale platforms.
| product type | CPU-GPU Superchip |
|---|---|
| Architecture / Series | NVIDIA Grace Blackwell GB200 |
| Processor Configuration | 1 Grace CPU / 2 Blackwell GPUs |
| Memory | 372 GB HBM3e + up to 480 GB LPDDR5X |
| Availability Status | Available / RFQ |
