NVIDIA DGX B200 10U AI system with 8 Blackwell GPUs, 1.44 TB HBM3e, NVLink, dual Xeon CPUs, ConnectX-7 networking, and BlueField-3 DPUs for training and inference.
Ground, Overnight
Fast Shipping
In Stock
100k+ SKUs
Sales and Support
8/5 Support
New / Refurbished
Certified
NVIDIA DGX B200 8-GPU Blackwell AI System | DGX-B200
⭐ Top Rated Verified Buyers
We are trusted for quality, reliability and fast shipping 💯
0
People watching this product now!
Description
NVIDIA DGX B200 8-GPU Blackwell AI System (DGX-B200)
NVIDIA DGX B200 is a 10U integrated Blackwell AI system with eight NVIDIA Blackwell GPUs interconnected through fifth-generation NVLink. It provides 1.44 TB of aggregate HBM3e memory, dual Intel Xeon host processors, ConnectX-7 scale-out networking, BlueField-3 DPUs, local NVMe storage, and the NVIDIA DGX software stack.
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 | DGX-B200 |
|---|---|
| Product Type | DGX 8-GPU AI System |
| GPU Configuration | 8x NVIDIA Blackwell GPUs |
| GPU Memory | 1,440 GB HBM3e total |
| GPU Memory Bandwidth | 64 TB/s aggregate |
| CPU | 2x Intel Xeon Platinum 8570 |
| System Memory | 2 TB standard; configurable to 4 TB |
| Networking | 8x ConnectX-7 VPI up to 400 Gb/s; 2x BlueField-3 DPUs up to 400 Gb/s |
| Form Factor / Power | 10U / approximately 14.3 kW max |
| Availability Status | Available / RFQ |
Deployment and Compatibility Considerations
- Validate rack power, airflow, operating temperature, service clearance, and PDU design for a 10U system drawing up to roughly 14.3 kW.
- Choose InfiniBand or Ethernet scale-out fabric and optics to match ConnectX-7 and cluster requirements.
- Size local and external storage for datasets, checkpoints, model artifacts, and inference context.
- Plan DGX OS, Mission Control, AI Enterprise, support, and cluster management with the intended deployment.
Common Use Cases
- Enterprise AI training
- Generative AI inference
- Fine-tuning and post-training
- DGX BasePOD and SuperPOD deployments
Frequently Paired Infrastructure
- 400G Spectrum-X or Quantum InfiniBand fabrics
- BlueField-3 DPUs
- AI storage
- OSFP/QSFP112 optics and LinkX cables
- Rack PDUs and data center cooling
Official NVIDIA Documentation
NVIDIA DGX B200 Specifications
Frequently Asked Questions
How many GPUs are in DGX B200?
DGX B200 contains eight NVIDIA Blackwell GPUs.
How much total GPU memory does DGX B200 provide?
According to NVIDIA: 1,440 GB, or 1.44 TB, of aggregate HBM3e GPU memory.
Additional information
| product type | GPU Server / DGX AI System |
|---|---|
| Platform / Architecture | NVIDIA DGX B200 / Blackwell |
| Accelerator Configuration | 8 Blackwell GPUs / 1.44 TB HBM3e |
| Form Factor / Power | 10U / ~14.3 kW max |
| Availability Status | Available / RFQ |
About brand
NVIDIA is the leader in GPU computing and AI infrastructure for high-performance workloads.
