Test report DSG-1682 · Rev E · tested November 18, 2026

Supply Chain & PolicyDevice under test

Nvidia Locked In $111 Billion More Memory Supply in One Quarter

Nvidia added $111 billion in supply commitments in one quarter, locking in years of memory as it raised its FY2028 growth forecast to 75% after Jensen Huang's exit.

Read
3 min
Words
501
Node
5nm
Operator
Elena Vasquez

Spec summary

  1. Nvidia's supply commitments rose $111 billion in one quarter, to $390 billion as of Oct. 25.
  2. Vera Rubin generated about $20 billion in its first volume-shipment quarter, ~19% of Q3 data center revenue.
  3. Nvidia raised its fiscal 2028 revenue growth forecast from 70% to 75%.
  4. Interim CEO Colette Kress confirmed secured memory supply; TSMC executives visited Nvidia this week.
Nvidia Secures Memory Supply as Huang Exits, Lifts Growth Outlook
Fig. ANvidia Secures Memory Supply as Huang Exits, Lifts Growth Outlook — Brecht Corbeel / unsplash

Nvidia's supply commitments jumped by $111 billion (about 147 trillion won) in a single quarter, as the company locked in the memory it needs to keep AI chips flowing after founder Jensen Huang stepped down. That's the headline number from Nvidia's first earnings report since the leadership change, and it signals that relationships with key memory suppliers — including Samsung Electronics and SK hynix — remain stable through the transition.

The company had been building long-term memory procurement well before Huang's exit. Regulatory filings show commitments for component purchases and manufacturing capacity grew roughly 40% to $390 billion (about 521 trillion won) as of October 25, the end of the third quarter. That's up from $279 billion (about 373 trillion won) at the close of the second quarter. Nvidia said multi-year memory commitments drove most of the increase.

To put that $111 billion jump in perspective: the single-quarter increase alone is more than the annual revenue of most Fortune 100 companies.

Why does memory supply matter so much?

Memory is the component that determines how quickly Nvidia can ship Vera Rubin, its next-generation AI computing system. The platform pairs sixth-generation high-bandwidth memory (HBM4) in the GPU with LPDDR5X low-power DRAM in the CPU. No memory, no shipments — it's that simple for practitioners watching server delivery timelines.

Vera Rubin is already a material business, not a roadmap promise. In its first quarter of volume shipments, the system generated about $20 billion in revenue. That works out to roughly 19% of third-quarter data center revenue.

What changed in the growth forecast?

With the memory pipeline secured, Nvidia raised its revenue growth forecast for fiscal 2028 — next year — to 75%, up from the previous 70%. That five-point bump translates to a substantial amount of additional revenue on a base of Nvidia's size.

Interim Chief Executive Colette Kress was direct about the reasoning. "Demand continues to grow, and we have secured additional supply to deliver to customers," Kress said. "Given current demand and the additional supply, we are confident we can achieve 75% growth."

What did Kress say about Huang's exit?

Asked on the third-quarter earnings call on the 17th about the impact of Huang's resignation on the business, Kress pointed to supplier engagement as evidence of continuity. TSMC executives visited Nvidia's offices this week, she said, and the company has also secured memory and advanced packaging capacity.

For buyers and system integrators, the practical readout is this:

  • Memory for AI chip production is contractually locked in, spanning several years.
  • Vera Rubin volume shipments are running at a $20 billion quarterly pace and ramping.
  • Supplier relationships — Samsung, SK hynix, TSMC — survived the CEO transition intact.
  • Next year's growth guidance moved up, not down, despite the leadership change.

The market is reading the numbers as confirmation that Huang's departure hasn't disrupted the supply chain. Nvidia's procurement machine, in short, kept running through the handover.

via en.sedaily.com (Source)

Filed under

  • nvidia
  • ai-memory-supply
  • hbm4
  • vera-rubin
  • memory-suppliers
Share this article:

More from Elena Vasquez

Elena Vasquez

Show full bio

Senior reporter covering people, launches and day-to-day practice in hardware layer of artificial intelligence at Die Signal.

675 articles

Same lot · LOT-362E

« Previous articleNext article »