Skip to main content
Invest AI
From market noise to cited evidence

Research context · not a recommendation

Broadcom (AVGO)

Descriptive positioning only. The positioning label says whether a company's business is exposed to a layer, never whether to buy it. Layers with no exposure are not recommendations to add any.

Stack position

Where AVGO sits (3 layers)

  1. L4 · Compute

    GPUs, AI accelerators, and general-purpose compute silicon for training and inference.

  2. L4 · HBM / Memory

    HBM/DRAM bandwidth beside accelerators, plus enterprise NAND/HDD capacity. US-listed tickers or ADRs only.

  3. L4 · Custom Silicon

    ASICs and custom accelerator designs (in-house or merchant silicon) for specific AI workloads.

  4. L5 · Networking

    Interconnect, switching, and high-speed data-movement silicon linking compute clusters.

Icons mark layer roles — not company logos. Position only; not a price chart.

AVGO appears on 3 layers because its primary businesses span more than one constraint — not a ranking. See Method for Broadcom / Marvell / Vertiv. Why multi-layer names

L4 · Compute

XingAI taxonomy. Filers did not assign this layer.

Map role vs this layer: Primary business sits here. Describes where the business sits, not a trade.

Broadcom

Products on file: Custom ASICs

What does Broadcom do in this layer?

Designs customized system-on-chip ASICs for customers' own systems, including AI accelerators.

Why does this layer matter for AI demand?

The visible centre of the chain, and the layer most people stop at. It is also the layer that cannot run alone: an accelerator needs memory bandwidth, packaging, interconnect, optics, storage, power and cooling around it. Any one of those lagging and the silicon idles.

What could change the picture here?

Design leadership concentrates here, but the near-term ceiling is usually set elsewhere — packaging, HBM supply, or power — rather than by accelerator design.

Same-layer companies (alphabetical, not ranked)

AMD ARM CBRS INTC NVDA

  • AMD — Supplies Instinct data-center GPUs used for AI training and inference.
  • ARM — Licenses CPU architecture and Neoverse designs that partners build into data-center and cloud AI chips.
  • CBRS — Builds wafer-scale AI systems, currently marketed for fast inference.
  • INTC — Supplies Xeon data-center CPUs used for AI, HPC and general-purpose workloads, and Gaudi AI accelerators.
  • NVDA — Supplies data-center GPUs and rack-scale systems for AI training and inference.

Open the Compute layer

L4 · Custom Silicon

XingAI taxonomy. Filers did not assign this layer.

Map role vs this layer: Primary business sits here. Describes where the business sits, not a trade.

Why does this layer matter for AI demand?

Hyperscalers large enough to amortise a tape-out design their own accelerators rather than buying only merchant parts. That shifts value toward the design partners and IP that make an in-house chip possible.

What could change the picture here?

Custom silicon competes for the same advanced packaging and HBM supply as merchant accelerators. Designing your own chip does not create capacity, it reallocates it.

Custom ASIC design partner for hyperscaler in-house accelerators.

Same-layer companies (alphabetical, not ranked)

MRVL

Open the Custom Silicon layer

L5 · Networking

XingAI taxonomy. Filers did not assign this layer.

Map role vs this layer: Primary business sits here. Describes where the business sits, not a trade.

Why does this layer matter for AI demand?

A frontier training run spans tens of thousands of accelerators that must behave as one machine. That produces a distinct hierarchy: links inside the package, scale-up inside the rack, scale-out between racks, plus the switches and NICs that carry it. This is a first-class industry, not a footnote under semiconductors.

What could change the picture here?

Cluster scale is increasingly limited by what the fabric can sustain rather than by how many accelerators can be bought.

Merchant switch silicon and datacenter networking components.

Same-layer companies (alphabetical, not ranked)

ALAB ANET APH CRDO CSCO MRVL

Open the Networking layer

Weekly digest: AI-map layer changes and new public 13F filings, each with its source. No score, no grade, no trade signal. Unsubscribe anytime.

Cache snapshot 2026-10-20. Filing dates below are the source dates, not this snapshot.

Public 13F rows

Rows below are public SEC Form 13F-HR filings, reported by the filing entity for the period shown and filed later — they are never current holdings. Form 13F covers long US equity positions only: it excludes short positions, cash, non-US holdings and most derivatives, and a reported position may have been sold since. Assigning a ticker to a layer is XingAI's classification, not the filer's. Nothing here is a recommendation, and an institution reporting a position is not a reason to take one.

What this page does not answer

Valuation, a target price, and when to buy or sell. Those are not in the cache, and this page will not guess them.

Research questions

  • Which layer is this company's primary business in, on XingAI's taxonomy?
  • Which public Form 13F-HR rows cite this ticker, and when were they filed?
  • Who else in the same layer is listed, alphabetically — not ranked?

Weekly digest: AI-map layer changes and new public 13F filings, each with its source. No score, no grade, no trade signal. Unsubscribe anytime.

    Broadcom (AVGO) · AI supply chain · Compute · XingAI Invest AI