Computing Interconnects

InfiniBand (NVIDIA Quantum)

Critical87% confidenceProfile verified Jul 13, 2026

A specialized low-latency networking technology used to connect GPUs across racks and clusters at scale, with NVIDIA's Quantum-X800 platform delivering 800Gb/s per port for large AI training clusters. Microsoft Azure, Oracle Cloud Infrastructure and CoreWeave are among the earliest adopters of NVIDIA's X800 series of 800Gb/s InfiniBand and Ethernet switches.

Sources

Every claim traced to a primary source — evidence, recent activity, and insider filing behavior, all in one place.

Evidence

2 sources · 87% avg. source confidence

  • NVIDIANVIDIA Quantum-X800 InfiniBand delivers 800Gb/s per port for trillion-parameter-scale AI clusters88% source confidence
  • NVIDIA NewsroomMicrosoft Azure and CoreWeave among early adopters of NVIDIA's X800 networking platforms85% source confidence

Dependency Map

What InfiniBand (NVIDIA Quantum) depends on below, and who depends on InfiniBand (NVIDIA Quantum) above — click any node to make it the new center, 2 levels deep.

Connection type

Investment
Dependency
Supply Chain
Development
Enablement
Competition / Other
Succession

Click any node to make it the new center. Scroll to zoom, drag to pan.

The Story So Far

Mar 2024: NVIDIA announced its X800 series of 800Gb/s networking switches, including the Quantum-X8…Mar 2024: Microsoft Azure, Oracle Cloud Infrastructure, and CoreWeave were named among the earliest…Mar 2024: AI data centers scaled toward trillion-parameter model training using NVIDIA's Quantum-X8…Mar 2024: Microsoft Azure and CoreWeave among early adopters of NVIDIA's X800 networking platformsJan 2026: Quantum-X800 platform ships

Auto-generated from this entity's dated milestones, relationship updates, and sourced evidence — not AI-written, just sorted.

InfiniBand (NVIDIA Quantum)'s Timeline

A sourced, dated history of InfiniBand (NVIDIA Quantum)'s key moments — founding to present.

  1. Jan 2026 · Quantum-X800 platform ships

    NVIDIA's Quantum-X800 InfiniBand platform began delivering 800Gb/s per port for trillion-parameter-scale AI training clusters, with Microsoft Azure, Oracle Cloud and CoreWeave among the earliest adopters.

How These Connections Evolved

How a relationship changed over time, not just its current state — sourced, dated, and traceable.

InfiniBand (NVIDIA Quantum)NVIDIA

  1. Mar 2024

    NVIDIA announced its X800 series of 800Gb/s networking switches, including the Quantum-X800 InfiniBand platform with 144 ports of 800Gb/s connectivity — double the prior Quantum-2 generation's per-port bandwidth.

  2. Now

    NVIDIA's Quantum InfiniBand platform, including the Quantum-X800 switch line, is developed by NVIDIA following its acquisition of Mellanox.

InfiniBand (NVIDIA Quantum)AI Data Centers

  1. Mar 2024

    Microsoft Azure, Oracle Cloud Infrastructure, and CoreWeave were named among the earliest adopters of NVIDIA's 800Gb/s X800 InfiniBand and Ethernet switches for trillion-parameter-scale AI training clusters.

  2. Now

    InfiniBand networking enables AI data centers to connect GPU clusters across racks at the scale required for trillion-parameter model training.

InfiniBand (NVIDIA Quantum)AI Data Centers

  1. Mar 2024

    AI data centers scaled toward trillion-parameter model training using NVIDIA's Quantum-X800 InfiniBand (800Gb/s per port), with hyperscalers including Azure, OCI, and CoreWeave among the earliest adopters.

  2. Now

    AI data centers depend on InfiniBand networking to connect GPU clusters across racks at the scale required for large-scale training.