At $162, Nvidia’s Going to Get Institutional Support From Those Who Missed Out Earlier, So I’m Buying

AI & Compute

Nvidia at $162.85: Why One Buyer Keeps Adding After a 26% Drop

Nvidia fell 26.41% in a month, yet Q2 revenue hit $96.22 billion and Vera Rubin ships to every major hyperscaler. One buyer keeps adding at $162.85.

By
Grace Kim
Filed
Channel
AI & Compute
Read
3 min read

Nvidia shares fell 26.41% in a month, and one investor at 24/7 Wall St. bought again at $162.85. The stock has been added on every leg down this fall. The stated reason is simple. The business underneath keeps getting stronger.

The thesis rests on Nvidia selling an entire AI factory, not just chips. Graphics processors, CUDA software, NVLink, networking and now CPUs ship as one platform. Customers who build on it return for the next generation. Vera Rubin began production shipments in August and holds purchase orders from every major hyperscaler.

Management projects fiscal 2028 revenue growth of approximately 70%. Supply, not demand, limits that number. On the August earnings call, Chief Executive Jensen Huang said: "Our demand is much higher than that."

What do the numbers show?

Growth keeps compounding. Q2 FY27 revenue reached $96.22 billion, up 105.8% year over year. Non-GAAP earnings per share of $2.22 beat the $2.09 consensus. That result marked six consecutive quarters of beats. Q3 guidance calls for $108.0 billion, and that figure includes no China data center compute.

Profitability leads the sector. Nvidia posts a 55.60% net margin and a 101.5% return on equity. Interest coverage stands at 503 times. Fiscal 2026 free cash flow came in at $96.58 billion.

Shareholders also get paid directly:

  • The quarterly dividend rose from $0.01 to $0.25.
  • Roughly $99 billion of buyback authorization remains.

Why the multiple matters now

The valuation has contracted. The trailing price-to-earnings ratio stands at 33. On forward estimates, the multiple has compressed into the low-to-mid 20s. That compression gives long-only funds that missed the earlier run a lower entry point to evaluate.

Some are already moving. Three of five tracked filers, including Marshall Wace and ARK Investment Management, added shares in the September quarter. The investor argues this institutional support should cushion the stock near current levels.

Why not AMD or Broadcom?

Advanced Micro Devices grows fast but trades at 202 times earnings on a 12.5% net margin. Its Q3 free cash flow came to $554 million, down 70.86%. Inventory rose 35% sequentially.

Broadcom posted AI semiconductor revenue of $16.70 billion. Even so, its quarterly free cash flow of $13.665 billion trails Nvidia's $21.34 billion. Broadcom lists its debt load and its reliance on a limited number of AI customers among its key risks.

What risks remain?

A leadership crisis concerns the buyer. A board investigation and reports about Huang's future sent the stock lower. Two major investors urged the board to consider his potential return.

Memory costs will also pull gross margins down to 71% to 72% in Q4. Supply commitments have reached $279 billion.

Those risks have left the thesis intact, because the demand belongs to customers rather than to any single executive. Cloud backlog runs above $2 trillion. Amazon Web Services is installing an additional 2 million GPUs through fiscal 2029.

The platform, in this view, outlasts any boardroom fight. CUDA, NVLink and the Vera Rubin roadmap will survive the current dispute.

What happens into earnings?

The next earnings report is close at hand, and the buyer plans to watch the stock around it. Revenue per gigawatt rises from $25 billion on Grace Blackwell to $40 billion on Vera Rubin. Every build-out cycle hands Nvidia a bigger check.

The position stays open because every dollar of AI infrastructure spending still runs through the company, and the author remains a buyer into the print on that basis.

Source: 247wallst.com

Share this article:

More from Grace Kim

Grace Kim

Show full bio

Market editor covering people, launches and day-to-day practice in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.

1,035 articles

Related articles

« Previous articleNext article »