Skip to content
Silicon Showcase
CompaniesApplicationsTechnology AreasAbout
Submit Technology
Submit Technology
Silicon Showcase

Silicon Showcase™ is the global directory of semiconductor technologies, platforms and innovations from around the world.

  • Technologies
  • Technology Signals
  • Companies
  • Applications
  • Technology Areas
  • About Us
  • How It Works
  • Submit Technology
  • Contact Us
  • Privacy Policy
  • Terms of Use
  • Disclaimer

Explore

  • Technologies
  • Technology Signals
  • Companies
  • Applications
  • Technology Areas

About

  • About Us
  • How It Works
  • Submit Technology
  • Contact Us

Legal

  • Privacy Policy
  • Terms of Use
  • Disclaimer

© 2026 Silicon Showcase™. All rights reserved.

Explore. Compare. Innovate.

Silicon Showcase
CompaniesApplicationsTechnology AreasAbout
Submit Technology
Submit Technology
Silicon Showcase

Silicon Showcase™ is the global directory of semiconductor technologies, platforms and innovations from around the world.

  • Technologies
  • Technology Signals
  • Companies
  • Applications
  • Technology Areas
  • About Us
  • How It Works
  • Submit Technology
  • Contact Us
  • Privacy Policy
  • Terms of Use
  • Disclaimer

Explore

  • Technologies
  • Technology Signals
  • Companies
  • Applications
  • Technology Areas

About

  • About Us
  • How It Works
  • Submit Technology
  • Contact Us

Legal

  • Privacy Policy
  • Terms of Use
  • Disclaimer

© 2026 Silicon Showcase™. All rights reserved.

Explore. Compare. Innovate.

  1. Home
  2. /Technologies
  3. /NVIDIA NVLink

HIGH-BANDWIDTH GPU-TO-GPU INTERCONNECT

High-bandwidth GPU-to-GPU interconnect within a server

RACK-SCALE ALL-TO-ALL FABRIC

Rack-scale all-to-all fabric when used with NVLink Switch

APPLICATIONS

AI & HPC · Data Center

Official technology image

HIGH-BANDWIDTH GPU-TO-GPU INTERCONNECT

High-bandwidth GPU-to-GPU interconnect within a server

RACK-SCALE ALL-TO-ALL FABRIC

Rack-scale all-to-all fabric when used with NVLink Switch

APPLICATIONS

AI & HPC · Data Center

Public source

NVIDIA NVLink

Compute & Accelerators

NVIDIA NVLink is a publicly named high-bandwidth GPU-to-GPU interconnect used to scale multi-GPU input and output inside a server and, with NVLink Switch, across a rack.

  • AI & HPC
  • Data Center
United StatesView Official Source (opens in a new tab)

Official technology image

Overview

NVIDIA NVLink is a high-bandwidth GPU-to-GPU interconnect and rack-scale fabric for AI training and inference. Distinct from NVLink-C2C chip-to-chip packaging links.

Official NVIDIA pages present NVLink as a scale-up fabric for AI training, inference and related GPU-accelerated workloads, including all-to-all GPU communication. NVLink Switch is a capability of this fabric rather than a separate listing. GPU architecture generations such as Hopper, Blackwell or Rubin are not independent interconnect technologies. NVLink-C2C, listed separately, is the package-level chip-to-chip coherent link, not this GPU-to-GPU fabric. Bandwidth figures and generation mappings should be confirmed on NVIDIA’s current NVLink pages.

Capabilities

Key characteristics associated with this technology.

  • High-bandwidth GPU-to-GPU interconnect within a server

  • Rack-scale all-to-all fabric when used with NVLink Switch

  • Described for AI training, inference and GPU-accelerated workloads

  • Distinct from NVLink-C2C package-level chip-to-chip interconnect

Organization

NVIDIA

United States

View organization

Official resources

  • NVIDIA NVLink and NVLink Switchnvidia.com (opens in a new tab)

Source & disclosure

Independent discovery listing. This profile was prepared from publicly available information published by the organization. For the latest and complete technical information, refer to the official company source.

Official source

nvidia.com (opens in a new tab)

Last source check

Aug 31, 2026

Notice something that needs updating?

Report a correction

Related Technologies

Explore other semiconductor technologies.

Compute & Accelerators

SenseSoC

Ultra-low-power connected-device system-on-chip that Sensesemi describes with wireless connectivity, a microcontroller, and an edge AI accelerator.

Consumer ElectronicsMedicalView Technology
Advanced Packaging

APTURA Thermo-Compression Bonding

Kulicke & Soffa APTURA is a thermo-compression bonder for advanced packaging, not a wire bonder catalog and not OSAT assembly.

AI & HPCData CenterView Technology
Materials & Substrates

Advanced IC Photomasks

Photronics advanced IC photomasks are lithography reticles for integrated circuits, not a foundry node and not FPD masks.

AutomotiveAI & HPCView Technology
Power Semiconductors

AlphaSGT

AlphaSGT is Alpha and Omega’s shielded-gate trench MOSFET technology, not a GaN HEMT and not a power-module substrate.

AI & HPCData CenterView Technology