NVIDIA, Broadcom, Qualcomm: Three AI Revenue Numbers, Three Different Clocks

The direct answer: do not rank these three numbers on one chart. NVIDIA's $75.2B and Broadcom's $10.8B are reported revenue for one quarter, but they cover different product boundaries. Qualcomm's more-than-$15B figure is an annual FY2029 company target, not current reported revenue.

The useful result is not a winner. It is three separate research lanes: NVIDIA for current platform scale, Broadcom for AI semiconductor momentum, and Qualcomm for whether a future target begins converting into a reported business.

Three big numbers, three different clocks

The comparison breaks before the arithmetic begins:

Use the five-label AI revenue field guide for the underlying method. Here, the labels change the question each company must answer.

Open the three source boxes (Source Evidence Snapshot)

NVIDIA: a broad business already reporting at scale

NVIDIA reported $75.2B of Q1 FY2027 Data Center revenue, up 92% year over year. Under its previous sub-market lens, compute was $60.4B and networking was $14.8B. That is the strongest current revenue evidence in this comparison, though the boundary is much broader than a single AI chip category.

Broadcom: a narrower AI total with hidden mix

Broadcom reported $10.8B of Q2 FY2026 AI semiconductor revenue, up 143% year over year, within $22.187B of total company revenue. The AI figure equals about 48.7% of total revenue. It is realized and material, but Broadcom does not split custom accelerators from AI networking in that total.

Qualcomm: a destination that still needs a reported path

Qualcomm's 2026 Investor Day release targets more than $15B of annual Data Center revenue in FY2029. Its business-update presentation shows a management ramp of $0.3B for FY2026e, $5B for FY2027, and $15B+ for FY2029.

The size of the ramp makes milestones more important, not less. FY2027 is 16.7 times the FY2026e figure, and the FY2029 target is at least 50 times that starting point. Those ratios describe the company's proposed path; they do not validate it.

Compare within a lane, not across the road

Once the labels are attached, the defensible comparison becomes a set of matched questions.

Editorial comparison visual. The lanes prevent a broad platform, a semiconductor subset, and a future target from becoming a false leaderboard.

This does not make NVIDIA automatically the best investment or Qualcomm automatically the weakest. A reported business can be priced aggressively; a target can create value before reaching full scale. The labels only tell us what is observable today and where the burden of proof sits.

Why annualizing does not repair the comparison

Multiplying the NVIDIA or Broadcom quarter by four would create an article estimate, not a company-reported annual result. It would also leave the boundary mismatch untouched. NVIDIA includes a broad Data Center platform; Broadcom reports AI semiconductors; Qualcomm presents a future annual destination.

The correct small-multiple view therefore keeps each lane on its own axis:

• NVIDIA: reported platform scale → reporting continuity → product and margin mix;
• Broadcom: reported AI semiconductor total → next-quarter delivery → product and margin mix;
• Qualcomm: management target → dated commercial milestones → reported revenue and economics.

For the deeper company paths, pair the NVIDIA reporting-lens review with the Blackwell cash-conversion test, then see the Broadcom AI mix review and Qualcomm expectation map.

Where even this comparison can mislead

The strongest counterargument is that careful labels can create a false sense of precision. The three businesses occupy different layers of AI infrastructure, and disclosure still omits key economics.

• NVIDIA's new customer-market lens may improve demand visibility while reducing continuity with the old compute/networking view.
• Broadcom's aggregate AI number may grow while its product mix, customer concentration, or margin quality changes.
• Qualcomm can hit intermediate revenue milestones yet deliver weaker economics than the headline target implies.
• None of these source lines alone settles valuation, cash conversion, capital intensity, or expected return.

The comparison is a research map, not a portfolio ranking.

Wait for the next matched checkpoint

The comparison should change only when the evidence changes. NVIDIA needs a usable bridge under its new labels. Broadcom's $16.0B Q3 FY2026 guidance needs a reported result with the same boundary. Qualcomm's $0.3B → $5B → $15B+ ramp needs reported commercial milestones.

Editorial monitoring visual using the official company disclosures. It records falsifiable follow-ups rather than predicting results.

Source-derived handoff: each gate stays inside the company's disclosed period and business boundary.

Until those checkpoints arrive, the honest summary is compact: NVIDIA shows current scale, Broadcom shows reported AI semiconductor momentum, and Qualcomm shows a target that still has to become reported revenue.

The next adjacent investigation is AMD's Data Center profit test, which asks whether reported segment growth becomes operating income and cash rather than relabeling it as product-level AI revenue.

The memory-supply counterpart is Micron's price-and-contract cycle test, which separates shipment growth from pricing and asks whether take-or-pay floors outweigh contract ceilings and new capacity. U.S. market access to another HBM supplier is examined in the SK hynix ADR parity test.

The physical-power handoff is Vistra's two-cash-calendar review, where AI demand moves from chip revenue labels into contracted electricity delivery, plant work, and cash conversion.

Method, sources, and disclosure

This comparison uses official company materials only. It preserves disclosed periods and boundaries, does not annualize quarterly figures, does not estimate undisclosed product mix, and does not treat targets as reported revenue. The 48.7% Broadcom share is $10.8B / $22.187B; Qualcomm's 16.7 times and at-least-50-times ratios use management's rounded framework.

• NVIDIA Q1 FY2027 results
• Broadcom Q2 FY2026 results
• Qualcomm 2026 Investor Day data-center strategy
• Qualcomm Investor Day 2026 business update, page 11

AI assisted with structure and consistency checks; final editorial responsibility remains with Hynexly's owner-operated desk. No sponsorship or affiliate relationship with the cited companies is disclosed. This is general information, not individualized investment advice; it does not issue a rating or share-price objective.