NVOnline #1151654
Compute / cluster
NVIDIA Vera Rubin NVL72 Reference Design
Generation-specific AI factory pack for Vera Rubin NVL72. NVIDIA lists it on the public DSX hub as NVOnline-gated.
NP-DSX/NVIDIA DSX
How we read NVIDIA's campus docs. We do not build that campus.
When a hall partner or a customer program needs DSX language, we restate the public overview to frame the hall's site, power, cooling, and compute-hall constraints. The 157-acre, 250 MW example is NVIDIA's published sizing table, not a NeuronPlant delivery. We are not an NVIDIA Cloud Partner today, and we do not have NVOnline access. We do not republish those packs.
Named packs
NVIDIA lists these on the public DSX hub. Full files sit behind NVOnline. Titles and document numbers only. No drawings, no BOM, no copied pages.
NVOnline #1145739
Site / facilities
NVIDIA DSX Vera Rubin Facilities Infrastructure Reference Design
Facilities pack for the Vera Rubin generation. Same NVOnline gate. We do not reproduce the drawings or BOM.
NVOnline #1152370
Campus design guide
NVIDIA DSX Facilities Infrastructure Design Guide
NVIDIA's public overview cites Design Guide v2.0 (19 August 2026) as the rendering source. The full guide is NVOnline.
NP-CAMPUS/Public overview
What the open DSX campus actually contains.
Restated from NVIDIA's DSX Facilities Infrastructure Reference Design Overview. That page is written for NVIDIA Cloud Partners. We use it as a map of the problem a hall partner or customer program may need to speak, not as a license to ship NVIDIA's design, and not as a campus NeuronPlant constructs.
NP-DSX / Campus stack
NVIDIA's map. Not a NeuronPlant campus
- 01
Grid / substation
Utility lands above 100 kV, steps into a 34.5 kV campus backbone with two buses: Network Core and GPU Compute.
- 02
BESS and standby generation
BESS on the 34.5 kV bus as a buffer for spiky AI load. Standby generation in N+2 for Core continuity. Sizing is site-specific.
- 03
Central Utility Building
Shared plant for liquid and air. Chillers on 4.16 kV. Pumps, dry coolers and controls on 480 V. UPS only on the Core side of the CUB.
- 04
Dry coolers
Air-side heat rejection, no evaporative water in this reference pattern. NVIDIA states a 45°C liquid-cooling design point so more of the facility budget stays with compute.
- 05
CDU gallery
Liquid-to-liquid CDUs in the mechanical gallery, N+1 groups. Secondary loop to GPU cold plates. NVIDIA publishes a TCS design flow of at least 1.5 LPM/kW.
- 06
Compute data hall
A Scalable Unit is one compute hot-aisle plus one support hot-aisle. Cabinet TDP in the public overview runs from 198 kW (MGX Gen 1.1) to 330 kW (Vera Rubin NVL72).
- 07
CIN spine and network core
East-West cluster interconnect with a 500 m optical reach limit. That number is a fiber plant constraint (SMF, MPO, polarity, test), not a switch SKU. Network Core holds tenant access (North-South), secure management (not application North-South), high-speed storage (its own data fabric when dedicated) and meet-me rooms.
NVIDIA published example, not a NeuronPlant project
250 MW-class IT load / 96 Scalable Units
NVIDIA presents this table as a sizing reference independent of the campus rendering. It is not a NeuronPlant project, not a campus we construct, and not a hall we sell. Actual sites differ. GPU rack count in that example assumes a maximum of eight racks per compute row.
- Site footprint
- 157 acres
- Building
- 972,500 SF
- Compute halls
- 4
- GPU racks
- 1,536
- GPUs
- 110,592
- Compute IT + mechanical
- 240 MW
- Core IT + mechanical
- 22.8 MW
What NeuronPlant does with this
Read the public campus so a hall partner or customer program can speak the same language: utility lead time, 34.5 kV topology, water vs dry rejection, CDU gallery, hall SU count, and whether Vera Rubin NVL72 at 330 kW cabinet TDP even fits the floor and the loop they already have, or the custom-system supplier they will use.
NeuronPlant does not pour that slab. We do not deliver NVIDIA's 157-acre example. When NVOnline access exists, work from the generation pack, not from this paraphrase. Until then, the public overview is the ceiling of what we will quote as NVIDIA.
NVIDIA, DSX, Vera Rubin and NVOnline are NVIDIA trademarks. NeuronPlant is independent of NVIDIA.
Sources: NVIDIA DSX Facilities Infrastructure Reference Design Overview; NVIDIA DSX documentation.
NP / Intake
Tell us what you need to achieve with AI.
One accountable delivery partner from AI requirement to a delivered factory: site, power, cooling, compute, fabric, structured cabling & passive infrastructure, storage, platform, procurement, and handover. You do not need a vendor list or a hall name first. The hall is a supplier.