NVIDIA GB300 NVL72 fully liquid-cooled platform with 72 Blackwell Ultra GPUs, 36 Grace CPUs, 20 TB GPU memory, and 130 TB/s NVLink for AI reasoning and inference.
Ground, Overnight
Fast Shipping
In Stock
100k+ SKUs
Sales and Support
8/5 Support
New / Refurbished
Certified
NVIDIA GB300 NVL72 Blackwell Ultra Rack-Scale AI Platform | GB300-NVL72
⭐ Top Rated Verified Buyers
We are trusted for quality, reliability and fast shipping 💯
0
People watching this product now!
Description
NVIDIA GB300 NVL72 Blackwell Ultra Rack-Scale AI Platform (GB300-NVL72)
NVIDIA GB300 NVL72 is the Blackwell Ultra generation of NVIDIA’s liquid-cooled NVL72 rack-scale architecture. It integrates 72 Blackwell Ultra GPUs and 36 Grace CPUs and is purpose-built for AI reasoning, test-time scaling inference, high-throughput agentic workloads, and large AI factory deployments.
Availability / RFQ: NVIDIA currently lists GB300 NVL72 as Available Now. Allocation, OEM system implementation, support, data center infrastructure requirements, and lead time still vary by project. 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 | GB300-NVL72 |
|---|---|
| Product Type | Blackwell Ultra Rack-Scale AI Platform |
| GPU / CPU Configuration | 72 Blackwell Ultra GPUs / 36 Grace CPUs |
| GPU Memory | 20 TB aggregate |
| GPU Memory Bandwidth | Up to 576 TB/s |
| CPU Memory | 17 TB LPDDR5X |
| NVLink Bandwidth | 130 TB/s |
| Cooling | Fully liquid-cooled rack-scale architecture |
| Scale-Out | Quantum-X800 InfiniBand or Spectrum-X Ethernet |
| Availability Status | Available / RFQ |
Deployment and Compatibility Considerations
- Plan liquid cooling, high-density rack power, CDU, piping, service space, and redundancy before deployment.
- Design scale-out fabric around ConnectX-8-class 800G connectivity and the chosen Ethernet or InfiniBand topology.
- Validate storage performance for reasoning, long-context inference, checkpointing, and training data.
- Confirm the exact OEM rack implementation and deployment/support package.
Common Use Cases
- AI reasoning and test-time scaling
- Agentic and long-context inference
- Large-model training
- Hyperscaler, neocloud, and sovereign AI factories
Frequently Paired Infrastructure
- Quantum-X800 or Spectrum-X Ethernet
- ConnectX-8 SuperNICs
- BlueField infrastructure
- AI-native storage
- Liquid-cooling and high-density power
Official NVIDIA Documentation
Frequently Asked Questions
Is GB300 NVL72 available?
NVIDIA currently labels GB300 NVL72 as Available Now.
How much GPU memory does GB300 NVL72 provide?
NVIDIA lists 20 TB of aggregate GPU memory in the rack-scale platform.
Additional information
| product type | Rack-Scale AI Platform |
|---|---|
| Architecture / Series | NVIDIA Blackwell Ultra GB300 |
| Accelerator Configuration | 72 Blackwell Ultra GPUs / 36 Grace CPUs |
| Cooling Type | Fully Liquid-Cooled Rack-Scale |
| Availability Status | Available / RFQ |
About brand
NVIDIA is the leader in GPU computing and AI infrastructure for high-performance workloads.
