| Key Features/Applications | Research and national lab workloads, AI development, AI inference and training, data science |
|---|---|
| CPU | NVIDIA GB300 Grace Blackwell Ultra Superchip with 72-core NVIDIA Grace CPU |
| Chassis | Tower or 5U Rackmount |
| Drive | 4 x M.2 PCIe 5.0 NVMe slots |
| RAM | 496 GB LPDDR5X SOCAMM |
| GPU | 1 x NVIDIA Blackwell Ultra GPU, 252 GB HBM3e |
| Network Ports | NVIDIA ConnectX-8 SuperNIC + 1 x 10GbE RJ45 + dedicated IPMI |
| Model | ARS-511GD-NB-LCC |
| Series | AI Workstation |
| Processor Family | NVIDIA Grace (Arm Neoverse) |
| Processor Count | 1 |
| Cores per Processor | 72 |
| Total System Memory | 496 GB LPDDR5X SOCAMM |
| Coherent Memory | 748 GB unified CPU and GPU |
| GPU Model | NVIDIA Blackwell Ultra |
| GPU Count | 1 |
| GPU Memory | 252 GB HBM3e |
| Boot Storage | 4 x M.2 PCIe 5.0 NVMe slots |
| Form Factor | Tower or 5U Rackmount |
| Standard Networking | NVIDIA ConnectX-8 SuperNIC, 10GbE RJ45, dedicated IPMI LAN |
| Cooling | Liquid cooled |
| Availability | Confirmed at quotation |
| Lead Time | Confirmed at quotation |
| Warranty | 3 Years Parts and Labour |
AI Workstation with NVIDIA GB300 Grace Blackwell Ultra (ARS-511GD-NB-LCC)
Starting at$259,945.04
In Stock
Confirmed at quotation
- Key Features/Applications: Research and national lab workloads, AI development, AI inference and training, data science
- CPU: NVIDIA GB300 Grace Blackwell Ultra Superchip with 72-core NVIDIA Grace CPU
- Chassis: Tower or 5U Rackmount
- Drive: 4 x M.2 PCIe 5.0 NVMe slots
- RAM: 496 GB LPDDR5X SOCAMM
- GPU: 1 x NVIDIA Blackwell Ultra GPU, 252 GB HBM3e
- Network Ports: NVIDIA ConnectX-8 SuperNIC + 1 x 10GbE RJ45 + dedicated IPMI
Request a quotation for volume pricing
Request a QuotationShipped Configuration
The base build as despatched. Every component group can be specified differently against your requirement, and the revised price is confirmed in a written quotation.
- CPUNVIDIA GB300 Grace Blackwell Ultra Superchip with 72-core NVIDIA Grace CPU
- Memory496 GB LPDDR5X SOCAMM
- Storage4 x M.2 PCIe 5.0 NVMe slots
- AOCNVIDIA ConnectX-8 SuperNIC + 1 x 10GbE RJ45 + dedicated IPMI
- GPU1 x NVIDIA Blackwell Ultra GPU, 252 GB HBM3e
Warranty and Onsite Service
3 Years Parts and Labour
Cover is the manufacturer warranty applicable to this product and region. Onsite service and extended terms can be added at quotation stage.
Configuration Summary
- CPUNVIDIA GB300 Grace Blackwell Ultra Superchip with 72-core NVIDIA Grace CPU
- Memory496 GB LPDDR5X SOCAMM
- Storage4 x M.2 PCIe 5.0 NVMe slots
- AOCNVIDIA ConnectX-8 SuperNIC + 1 x 10GbE RJ45 + dedicated IPMI
- GPU1 x NVIDIA Blackwell Ultra GPU, 252 GB HBM3e
- Warranty3 Years Parts and Labour
Need help specifying this system?
Rack power, cooling capacity and network topology all change what will actually work on your site. Tell us the environment and we will come back with a configuration that fits it.
Talk to a specialistThe ARS-511GD-NB-LCC puts a full NVIDIA Grace Blackwell Ultra node within reach of a research group, a model development team or a departmental lab, without the power and cooling profile of a rack-scale cluster. The Grace CPU and the Blackwell Ultra GPU share a single coherent memory space of 748 GB, so large models and large datasets stay resident rather than being paged across a PCIe boundary.
Liquid cooling keeps the system viable in a lab or office environment, and the chassis can be deployed as a tower on the floor or racked at 5U when the work moves into a data centre.
iVerified sources this platform through authorised channels, verifies the delivered configuration against your written specification, and supplies the serial-level documentation your asset and audit records require.




