Vera Rubin, delivered as a service.
Liquidata builds on NVIDIA's Vera Rubin platform and delivers it as dedicated, managed clusters on powered sites, with the network, storage, monitoring, security and operations around it.
One rack, seventy-two GPUs, one liquid loop.
The Vera Rubin NVL72 rack pairs 72 Rubin GPUs with 36 Vera CPUs over NVLink, cooled entirely by liquid with 45 °C inlet water, which allows cooling without chillers in most climates. NVIDIA reports the platform in full production since January 2026, with partner availability in the second half of 2026.
| Specification | Per NVL72 rack | Note |
|---|---|---|
| GPUs and CPUs | 72 Rubin GPUs, 36 Vera CPUs | One NVLink domain |
| NVFP4 inference | 3,600 PFLOPS | NVIDIA published figure |
| FP8 training | 1,260 PFLOPS | NVIDIA published figure |
| HBM4 memory | 20.7 TB, 1,400 TB/s | 288 GB per GPU |
| NVLink bandwidth | 216 TB/s | GPU to GPU |
| CPU memory | Up to 54 TB LPDDR5X | Vera CPUs |
| Cooling | 100% liquid, 45 °C inlet | Chiller-free in most climates |
| Rack power | About 190 kW | NVIDIA's standard setting; up to about 230 kW at maximum performance |
Size a module in racks. Read it in megawatts.
Modules are contracted in racks and delivered against a site's power. Move the slider to see what a module needs from the site and the grid.
190 kW per rack at NVIDIA's standard setting plus 5% for scale-out network and storage; 8,760 hours a year; 288 GB HBM4 per GPU.
The cluster, and everything that keeps it earning.
Cluster design and bring-up
Rack layout, power and cooling integration with the site, fabric topology, burn-in and acceptance against the customer's workloads.
High-speed fabric and storage
Scale-out networking and storage sized to the workload, delivered with the module and operated as part of the service.
Monitoring, diagnostics and remote operations
Liquidata's own monitoring software: cooling energy, availability, density and component health, with remote operations from day one.
Security from site to tenant
Physical, network and tenant isolation designed in, for customers whose data and models cannot leave their jurisdiction.
Round-the-clock operations
Operations, service levels and support carried by Liquidata as the provider of record.
Energy-aware scheduling
Suitable workloads scheduled around renewable output, storage and prices at sites that have them, without touching service levels for the rest.
Reserved, then built, then delivered on the site's date.
Commitments of 12 to 36 months. Capacity is reserved on take-or-pay terms, priced for the term and the module, with Liquidata as provider of record.
Modules are ordered against demand. No material hardware order is placed until reservations cover about 70% of a module's sellable GPU-hours, and no site is committed until power, permits and delivery dates are confirmed in signed agreements.
Delivery is timed to energization. Each module's delivery date is tied to its site's power delivery date, so customers reserve against a date they can plan around.
Rubin Ultra follows. The 2027 generation moves to racks of about 600 kW, the density at which the cooling Liquidata specializes in is the only option. Reservations can be placed for either generation.