Skip to content
AI supply chain term

NVLink

NVLink is NVIDIA's high-speed interconnect that links GPUs directly so many chips can act as one tightly-coupled machine.

What it means

NVLink is a fast, direct chip-to-chip interconnect NVIDIA uses to join GPUs (and CPUs) with far more bandwidth than standard PCIe, and NVSwitch extends it so many GPUs can all talk to each other at full speed. This creates a tightly-coupled 'scale-up' domain in which a group of GPUs behaves almost like one giant accelerator sharing memory. In the AI supply chain it defines the size and efficiency of the tightest compute domain — how many chips can be fused before you must fall back to slower data-center networking. It is a critical lever for training large models and for high-throughput inference, and it is central to NVIDIA's rack-scale systems. The reach and bandwidth of this scale-up fabric increasingly set the performance ceiling for a single AI node.

Why it matters to investors

NVLink is part of why NVIDIA now sells racks and systems, not just chips, capturing more value and deepening lock-in at the interconnect layer. Connectivity specialists that supply retimers and links around these systems are exposed to how large scale-up domains are allowed to grow.

Companies on this part of the chain

NVIDIANVDAAstera LabsALABCoreWeave, Inc.CRWVNebius Group N.V.NBIS

Named to show where the term sits in the AI supply chain — research, not advice, and never a recommendation to buy or sell.

Related terms

Scale-out vs scale-upConnecting many machines into one cluster over a network fabric.Superchip (GPU-CPU module)A superchip is a module that tightly fuses one or more GPUs with a CPU over a fast internal link so they act as a single, memory-coherent compute unit.Collective communication (all-reduce)All-reduce is the collective communication step in which every chip in a cluster combines its partial results and ends up with the same final total — the core exchange that keeps distributed AI training in sync.SerDesSerializer/Deserializer — the silicon that drives high-speed electrical links.

See NVLink in the live AI chain.

THE ENTITY maps every constraint onto one live model — which part is tight now, who owns it, and who gets squeezed when it moves. Plain-English reads you can check.

Explore the features
All glossary terms

THE ENTITY is an educational read on the AI supply chain — research, not investment advice. It explains how the chain works and who sits where, never price targets or buy/sell calls.