NVIDIA · Platform · Rack system

NVIDIA Vera Rubin NVL72

Exact Vera Rubin NVL72 reference specifications and AI deployment guidance, separated from FusionPointAI commercial inventory.

VRAM20700 GB

Memory technologyHBM4

Form factorRack scale · 72 GPU · Platform

InterfaceMGX NVL72

PowerNot listed

Current Reference only

Compare GPU

Technical diagrams

Neutral technical reference illustration for NVIDIA Vera Rubin NVL72; not a product photograph — Reference illustration; consult the verified specification table for exact hardware identity. This is not a product photograph. FusionPointAI neutral GPU reference illustration

Specifications

Field	Value
Manufacturer	NVIDIA
Exact SKU	Vera Rubin NVL72
Product family	NVIDIA Vera Rubin NVL72
Architecture	Rubin
Generation	Rubin
Entity type	Rack system
Lifecycle	Current
Launch date	Not listed
VRAM	20700 GB
Memory technology	HBM4
Memory bandwidth	1,580,000 GB/s
Power	Not listed
Interface	MGX NVL72
Form factor	Rack scale · 72 GPU · Platform
Interconnect	NVLink 6
MIG / partitioning	Not supported
Supported numeric formats	FP64, FP32, TF32, BF16, FP16, FP8, FP6, FP4, INT8
Software stack	cuda

AI workload suitability

LLM inference

High memory capacity supports large checkpoints and multi-GPU inference when runtime and topology are certified.

Fine-tuning and training

Suitable for limited fine-tuning and development when software support, cooling and memory capacity are verified.

Image, video and embeddings

FP16 acceleration supports common generative and vector workloads within the listed VRAM envelope.

Power and cooling

The manufacturer does not publish a directly comparable board-power value for this entity.

Multi-GPU use

NVLink 6 is the recorded interconnect path; topology must still be validated at node level.

Software ecosystem

cuda

Product timeline