Brand: Dell | Category: Servers
SKU: DEMC-XE9712-B200 | Part #: XE9712-B200 | MPN: XE9712-B200
Contact for Pricing — Request a Quote
Dell PowerEdge XE9712 — full-rack GB200 NVL72 system with 36× Grace CPU + 72× Blackwell GPUs delivering exascale AI.
The Dell PowerEdge XE9712 GB200 NVL72 Rack-Scale AI System (part number XE9712-B200) is a flagship enterprise AI server distributed worldwide by Omnixon Global. Built around the NVIDIA B200 accelerator family and engineered for production AI infrastructure, it is one of the most-requested platforms for foundation-model training, fine-tuning, and large-scale inference in 2025–2026.
Yes. Every unit ships, serial registration, and the official software/firmware stack.
Lead time depends on the configuration and current stock position. Submit an RFQ with your destination and our team will confirm availability for your order.
Yes. We routinely deliver 8-node, 16-node, and 32-node B200 clusters with NVIDIA Quantum-2 InfiniBand or Spectrum-X Ethernet fabrics, parallel storage (DDN, VAST, Weka, Pure FlashBlade//E), and full rack/power/cooling design. Send your training-cluster BOM and we'll respond with a complete priced design.
Omnixon Global engineers configure full AI clusters — networking, storage, cooling, racks, deployment.
Request a Quote → Or talk to an engineerOmnixon Global supplies enterprise IT hardware. Request a quote or contact our team for project design and Bill-of-Materials review.
| Brand | Dell |
| Category | Servers |
| SKU | DEMC-XE9712-B200 |
| Part Number | XE9712-B200 |
| Condition | New |
| Form Factor | Rack-Scale System (1× full rack) |
| Interface | NVMe |
| GPU Support | 72× NVIDIA B200 Blackwell (NVL72 configuration, NVLink-5 connected) |
| AI Optimized | Yes |
| Processor | 36× NVIDIA Grace ARM-based CPUs (Grace-Blackwell superchips) |
| Total HBM | 13.5TB HBM3e across 72 GPUs |
| GPU Interconnect | NVLink Switch System, 1.8 TB/s per GPU |
| AI Performance | ~1.4 exaFLOPS FP4 sparse / 720 petaFLOPS FP8 |
| Memory (LPDDR) | 17TB unified Grace LPDDR5X |
| Drive Bays | Rack-scale NVMe array (vendor-configurable) |
| Network | NDR/XDR InfiniBand or 800GbE Ethernet |
| Power Supply | ~120kW per rack, liquid-cooled |
| Cooling | Direct liquid cooling (DLC), rear-door heat exchanger compatible |
| Form Factor | Full 42U-48U rack |
The Dell PowerEdge XE9712 GB200 NVL72 Rack-Scale AI System is built for enterprise data-center workloads — virtualization (VMware, Proxmox, Nutanix), private cloud, database hosting, and AI/ML training. Its Rack-Scale System (1× full rack) form factor fits standard EIA-310 server racks and supports redundant PSUs and hot-swap drives common in production environments.
Key specifications for the Dell PowerEdge XE9712 GB200 NVL72 Rack-Scale AI System: form factor Rack-Scale System (1× full rack); 3 years support; new condition; gpu 72× NVIDIA B200 Blackwell (NVL72 configuration, NVLink-5 connected); cpu 36× NVIDIA Grace ARM-based CPUs (Grace-Blackwell superchips); total hbm 13.5TB HBM3e across 72 GPUs; gpu interconnect NVLink Switch System, 1.8 TB/s per GPU; ai performance ~1.4 exaFLOPS FP4 sparse / 720 petaFLOPS FP8; memory (lpddr) 17TB unified Grace LPDDR5X. Manufacturer part number XE9712-B200. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.