As of 7:12 p.m. ET on Friday, December 26, 2025, NVIDIA Corporation (NASDAQ: NVDA) was trading around $190.53 in after-hours activity, with roughly 139.5 million shares traded during the day.
Friday’s session unfolded in a quiet, post-holiday market that kept major U.S. indexes close to recent highs. The Dow slipped to 48,710.97, the S&P 500 to 6,929.94, and the Nasdaq Composite to 23,593.10, with overall exchange volume notably lighter than recent averages. [1]
For Nvidia stock investors, the late-week focus has narrowed to a familiar set of themes—AI infrastructure spending, export controls, and supply-chain strategy—but with new urgency after a non-exclusive AI inference technology licensing agreement with Groq and fresh developments around U.S.–China chip policy. [2]
Why Nvidia stock moved: the Groq partnership puts AI inference in the spotlight
The biggest headline driving near-term sentiment around NVDA stock is Nvidia’s deal with Groq, an AI-chip startup known for inference-focused technology. Groq said it entered a non-exclusive licensing agreement with Nvidia aimed at “accelerat[ing] AI inference at global scale,” and that Groq founder Jonathan Ross, Groq President Sunny Madra, and other team members will join Nvidia to help advance and scale the licensed technology. [3]
Reuters noted that Nvidia climbed in Friday’s session after agreeing to license Groq technology and hire its CEO—one of the clearest signals yet that Wall Street is rewarding Nvidia for extending its lead beyond AI training and into AI inference, where competition is heating up. [4]
The market debate isn’t whether inference matters—it does—but what inference economics look like as hyperscalers and rivals pursue specialized chips, custom silicon, and alternative architectures. Barron’s framed Nvidia’s Groq move as a strategic hedge as inference workloads surge, while also flagging the risk that broadening into inference-optimized approaches could pressure Nvidia’s premium profitability over time. [5]
Meanwhile, Investopedia reported that Nvidia’s shares were lifted by the Groq news and highlighted the confusion around deal structure circulating this week—emphasizing that Nvidia said it had not acquired Groq outright, even as market chatter focused on possible asset purchases and the strategic value of Groq’s inference capabilities. [6]
The core bull case remains: Nvidia’s AI revenue engine and guidance momentum
While deal headlines can move the stock day-to-day, Nvidia’s larger valuation debate is still anchored in two questions:
- How durable is AI infrastructure spending?
- Can Nvidia sustain margins and growth rates as competition rises?
Nvidia’s most recent quarterly report underscores why investors continue to treat NVDA as the bellwether for AI demand.
In its third quarter of fiscal 2026 (ended October 26, 2025), Nvidia reported:
- Revenue: $57.0 billion (up 22% quarter-over-quarter, up 62% year-over-year)
- Data Center revenue: $51.2 billion (up 25% quarter-over-quarter, up 66% year-over-year)
- GAAP gross margin: 73.4%
- GAAP EPS (diluted): $1.30 [7]
Just as important for the stock’s next leg: Nvidia’s outlook. The company guided for fourth-quarter fiscal 2026 revenue of $65.0 billion, plus or minus 2%, and projected GAAP gross margin of ~74.8% (±50 bps). [8]
That combination—rapid top-line growth and mid-70s gross margins—helps explain why Wall Street is still willing to assign Nvidia a market-leading multiple, even as investors debate how long AI capital expenditures can keep compounding at today’s pace. [9]
“AI bubble” fears vs. Nvidia’s rebuttal: bookings visibility and customer concentration
Nvidia has become the market’s scoreboard for AI. Reuters captured that dynamic in November, reporting that CEO Jensen Huang pushed back on “AI bubble” concerns and pointed to the breadth of Nvidia’s footprint across cloud and enterprise deployments. [10]
Two details from Reuters are especially relevant for investors weighing risk versus reward:
- Nvidia has referenced $500 billion in bookings through 2026 for advanced chips. [11]
- Nvidia’s revenue concentration is significant: Reuters reported that 61% of revenue came from four major customers, a reminder that hyperscaler purchasing behavior (and any shift toward in-house silicon) can matter enormously to the stock. [12]
Reuters also quoted Neil Azous, portfolio manager of the Monopoly ETF (which holds Nvidia), underscoring how Nvidia’s market role can amplify both upside and downside: when NVDA moves, it can influence sentiment well beyond semiconductors. [13]
This is exactly where the Groq deal fits in the narrative: it signals Nvidia wants to stay central as AI workloads evolve from training-heavy builds to inference-heavy production deployments.
China and export controls: a major NVDA catalyst heading into 2026
A second major driver for Nvidia stock forecasts right now is geopolitics—specifically, what the U.S. will allow Nvidia to ship into China, under what conditions, and how Beijing responds.
Trump administration: H200 exports to China with a 25% fee
In early December, Reuters reported that President Donald Trump said the U.S. would allow Nvidia’s H200 processors—described as its “second-best” AI chips—to be exported to China and collect a 25% fee on such sales. [14]
The same Reuters report highlighted the policy tightrope:
- U.S. officials balancing China access versus maintaining a global lead in AI chips
- National security concerns over advanced chips aiding China’s military
- The administration’s stance that Blackwell and Rubin are not part of the deal [15]
Reuters also included expert and analyst viewpoints illustrating the controversy:
- Eric Hirschhorn, a former senior Commerce Department official, warned against trading national security for trade advantage. [16]
- George Chen (The Asia Group) suggested Beijing could soften its posture depending on broader U.S.–China dynamics. [17]
- Bo Zhengyuan (Plenum) argued that longer-term, China remains focused on building domestic advanced chip capabilities even if near-term policy eases. [18]
Next step: shipments to China targeted for mid-February 2026—if approved
More recently, Reuters reported Nvidia told Chinese clients it aims to begin H200 shipments to China by mid-February 2026, with early shipments drawn from existing stock—5,000 to 10,000 chip modules, equivalent to roughly 40,000 to 80,000 H200 chips—but shipments are contingent on Beijing’s approval, according to sources. [19]
For NVDA investors, the key takeaway is that export policy can act like a lever on quarterly expectations—yet it’s not purely additive. A policy opening can increase unit opportunities, but it can also:
- raise compliance and political risk,
- add uncertainty around product segmentation (what can be sold where), and
- influence competitive dynamics with local Chinese suppliers.
Intel investment cleared, but foundry testing questions add nuance to Nvidia’s supply-chain story
Nvidia’s strategic positioning isn’t only about what chips it sells—it’s also about where the industry can manufacture at scale.
Reuters reported in December that U.S. antitrust agencies cleared Nvidia’s investment in Intel, citing an FTC notice; Nvidia previously announced a $5 billion investment in the chipmaker. [20]
But Reuters also reported a crucial wrinkle inside Intel’s foundry push: Nvidia tested whether it would manufacture chips using Intel’s 18A production process, but stopped moving forward, according to sources. [21]
That matters for Nvidia investors because it reinforces an often-overlooked reality: Nvidia’s AI roadmap depends not only on demand, but also on advanced manufacturing capacity and yields across the ecosystem.
Insider selling: Nvidia board member stock sale in focus
In mid-December, Reuters reported that Nvidia board member Harvey Jones sold more than $44 million worth of Nvidia shares, according to a regulatory filing. [22]
Insider sales can occur for many non-fundamental reasons (tax planning, diversification, scheduled programs), but large sales often draw attention because Nvidia remains such a widely held—and widely debated—mega-cap name.
Analyst forecasts and price targets: Wall Street stays constructive, but dispersion is real
Forecasts for Nvidia continue to skew bullish, and the Groq deal has reinforced the “Nvidia stays central to AI” narrative.
Investopedia reported that, as tracked by Visible Alpha, Nvidia’s mean price target is around $254, compared with recent prices around the low $190s. [23]
At the same time, the range of outcomes remains wide because the stock’s multiple is sensitive to:
- the pace of AI spending by hyperscalers,
- competition from custom inference silicon,
- export-control constraints, and
- margin sustainability as Nvidia extends its platform across more of the AI stack. [24]
Nvidia itself has continued to guide to mid-70s gross margins in its outlook—an important anchor for valuation models—yet the market will continue stress-testing whether new growth vectors (like inference partnerships) are accretive or dilutive to long-run margins. [25]
What investors should know before the next regular trading session
The regular U.S. stock market session is closed (Friday’s cash session has ended), and the next regular session is Monday, December 29, 2025.
Heading into Monday, Nvidia investors will likely focus on four practical items:
1) After-hours price moves can be noisy
With liquidity typically thinner after 4 p.m. ET—especially in the holiday stretch—price swings can look dramatic without signaling a true shift in institutional positioning. NVDA was around $190.53 in after-hours Friday evening.
2) Watch for follow-up disclosures on the Groq deal structure
Groq’s own statement is clear about non-exclusive licensing and team transitions. What investors will watch next are any additional details that clarify economics, integration plans, and how Nvidia intends to productize inference advances within its platform. [26]
3) China headlines remain “market-moving” for NVDA
Two separate milestones matter:
- U.S. policy implementation details around the H200 export framework [27]
- Whether Beijing approves imports and the timing/size of early mid-February 2026 shipments [28]
4) The next major scheduled catalyst: Nvidia’s Q4 FY26 results on February 25, 2026
Nvidia’s investor relations calendar lists its 4th Quarter FY26 Financial Results for February 25, 2026—a date that can influence positioning weeks in advance, particularly for a stock as widely owned as NVDA. [29]
Bottom line for NVDA stock
Nvidia enters the final trading days of 2025 with the same core investment premise—AI compute demand at scale—but with a shifting backdrop:
- The Groq licensing agreement signals Nvidia is leaning into the next phase of AI: inference at massive scale, where competitive threats are rising. [30]
- Export controls and China policy have re-emerged as near-term catalysts that can change quarterly narratives quickly. [31]
- Nvidia’s own results and outlook continue to show extraordinary growth and premium margins, keeping analysts constructive—even as the market debates customer concentration and longer-term margin durability. [32]
Going into Monday, the question isn’t whether Nvidia is still leading AI—it is. The question is whether new strategic moves, from inference partnerships to supply-chain and geopolitical navigation, can keep NVDA stock on a path where growth remains strong enough to justify its lofty expectations. [33]
References
1. www.reuters.com, 2. groq.com, 3. groq.com, 4. www.reuters.com, 5. www.barrons.com, 6. www.investopedia.com, 7. nvidianews.nvidia.com, 8. nvidianews.nvidia.com, 9. nvidianews.nvidia.com, 10. www.reuters.com, 11. www.reuters.com, 12. www.reuters.com, 13. www.reuters.com, 14. www.reuters.com, 15. www.reuters.com, 16. www.reuters.com, 17. www.reuters.com, 18. www.reuters.com, 19. www.reuters.com, 20. www.reuters.com, 21. www.reuters.com, 22. www.reuters.com, 23. www.investopedia.com, 24. www.barrons.com, 25. nvidianews.nvidia.com, 26. groq.com, 27. www.reuters.com, 28. www.reuters.com, 29. investor.nvidia.com, 30. groq.com, 31. www.reuters.com, 32. nvidianews.nvidia.com, 33. www.reuters.com
