# NVIDIA NVL72 **Entity class:** Rack-scale AI system **Domain:** Artificial intelligence / Compute **Maturity:** Developed ## Definition NVIDIA NVL72 is a rack-scale architecture that connects 72 Blackwell GPUs into a tightly coupled computing system for large-model training and inference. ## Mechanism and significance It treats networking, memory, power, cooling, and accelerators as one machine, shifting the unit of competition from an individual chip to an integrated AI factory. ## Relationships - **Research dossier:** [[research/ASI and RSI Timeline Research Moonshots|ASI and RSI Timeline Research Moonshots]] - **Ontology route:** [[ASI and RSI Timeline Ontology#Model Architecture and Compute Economics|Model Architecture and Compute Economics]] - **Primary fields:** [[Machine Intelligence]] · [[AI Infrastructure]] · [[Compute Race]] - **Adjacent concepts:** [[Organizational Intelligence Metabolism]] · [[NVIDIA B300]] · [[H100 Scarcity]] · [[GPU Secondary Market]] ## Sources and provenance - [[research/ASI and RSI Timeline Research Moonshots|ASI and RSI Timeline Research Moonshots]] — immediate source for this node's role in the broadcast research map. ## Evidence boundary The research dossier establishes why this entity or concept belongs in the Moonshots ontology. Time-sensitive organizational, product, policy, and performance claims should be checked against the linked primary source or a current authoritative source before reuse as settled fact. ## Simple Reminders, Quotations, and Thoughts > When frontier compute is scarce enough to be overbid, leased at extraordinary cost, or allocated through privileged relationships, scientific merit no longer determines who can run an experiment. Access politics becomes part of the research method. > **— Adapted from Dave Blundin**, *Moonshots, October 7, 2026* [[reminders/AI Infrastructure/Compute Scarcity Turns Scientific Merit Into Access Politics by Dave Blundin|Compute Scarcity Turns Scientific Merit Into Access Politics by Dave Blundin]]