Advanced Micro Devices, Inc. (NASDAQ: AMD) heads into the Friday, December 26, 2025 U.S. session with a familiar mix of powerful upside narratives—AI accelerators, hyperscaler partnerships, and an ambitious rack-scale roadmap—alongside one of the market’s biggest near-term swing factors: China-related export policy and demand visibility.
The U.S. stock market was closed on Christmas Day (Dec. 25) and had an early close on Dec. 24, setting up a potentially thin-liquidity re-open where headline momentum can move prices faster than usual. Nasdaq notes regular hours are 9:30 a.m. to 4:00 p.m. ET, with pre-market typically running 4:00 a.m. to 9:30 a.m. ET. [1]
Below is what to know before the bell—the latest AMD-specific headlines, the key product/partner catalysts driving 2026 expectations, the next earnings checkpoint, and where analysts’ forecasts sit right now.
AMD stock price snapshot into the Dec. 26 open
Because of the holiday schedule, the most recent trading reference is the Christmas Eve (Dec. 24) early-close session. Data providers differ slightly depending on whether they’re showing the regular close or extended-hours prints:
- One market snapshot shows AMD around $214.63 at the Dec. 24 close (1:00 p.m. ET). [2]
- Another shows AMD at about $215.04 with an intraday range roughly $214.00–$216.48 and volume near 8 million shares.
When markets reopen after a holiday, it’s common to see lighter volume and larger-than-usual gaps on breaking headlines—especially in high-beta semiconductors.
The headline risk (and potential upside): China, MI308 exports, and a reported Alibaba order
1) AMD’s CEO met China’s commerce minister in Beijing
Reuters reported that China’s Commerce Minister Wang Wentao met with AMD CEO Lisa Su in Beijing to discuss AMD’s business development in China and “strengthening cooperation.” The public statement didn’t include deal specifics, but the timing matters given the current export-control landscape. [3]
2) AMD says it has licenses—and is willing to pay a 15% fee—to ship MI308 to China
Earlier in December, Reuters reported Su said AMD has licenses to ship some MI308 chips to China and is prepared to pay a 15% fee/tax to the U.S. government if it ships them. Reuters also noted that the U.S. administration described this as a deal allowing AMD and Nvidia to resume shipping some chips to China in exchange for the fee—while some legal experts argued such a structure could raise constitutional questions. [4]
This is not a small detail for AMD’s near-term modeling: AMD itself has said its current outlook does not include revenue from MI308 shipments to China. That means any confirmed shipments could be viewed by the market as incremental upside relative to guidance—though policy and compliance risk remains. [5]
3) Report: Alibaba considering 40,000–50,000 MI308 accelerators
Multiple reports in late December put a sharper point on the “China demand” question. TechNode reported that Alibaba is considering purchasing 40,000 to 50,000 AMD MI308 AI accelerators, describing MI308 as tailored for the China market and requiring the 15% licensing fee; the report also cited specs such as 192GB of HBM3 and positioned pricing versus Nvidia’s China-focused alternatives. [6]
Separately, market coverage has linked AMD’s setup into year-end to the same reported Alibaba/MI308 narrative, with the broader takeaway being: if large China orders become real and shippable, investors may quickly re-rate near-term revenue optionality. [7]
What to watch pre-market: any confirmation (or denial) from AMD, Alibaba, or regulators. In chip exports, “reported order” and “recognized revenue” can be months apart, so the market’s reaction often depends on whether the story moves from rumor → licensing clarity → shipment timing.
The AI mega-deal that reset expectations: OpenAI + AMD (and what the timeline implies)
One reason AMD’s headline sensitivity is high is that the company has signed unusually large, visibility-creating partnerships.
OpenAI agreement: massive scale, but most of the compute starts in 2H 2026
Reuters reported that AMD will supply OpenAI with hundreds of thousands of AI chips (GPUs) over several years, describing the deployment scale as six gigawatts beginning in the second half of 2026. The deal also includes a warrant structure giving OpenAI the option to buy up to ~10% of AMD, and Reuters reported AMD executives expected the ripple effect of the agreement to drive more than $100 billion in new revenue over four years from OpenAI and other customers. [8]
AMD’s own announcement framed the partnership as expected to deliver tens of billions of dollars in revenue and said the agreement is expected to be highly accretive to non-GAAP EPS. [9]
Why this matters for Dec. 26 trading: the OpenAI deal is widely seen as a long-duration catalyst, but it also increases investor focus on whether AMD can execute on:
- next-gen accelerator ramps
- system-level delivery
- software ecosystem maturity
- supply chain and capacity planning
Those execution questions can drive short-term volatility even when the long-term narrative is bullish.
“Helios” and the rack-scale strategy: HPE adoption + Oracle deployments
AMD’s pitch to investors is increasingly system-level, not just “we sell chips.”
HPE: one of the first OEMs to adopt AMD “Helios,” available globally in 2026
AMD announced an expanded collaboration with HPE in which HPE will be among the first OEMs to adopt the AMD “Helios” rack-scale AI architecture, integrating networking (including Juniper networking switches) and collaboration with Broadcom for Ethernet-based scale-up connectivity. AMD also said HPE will offer Helios worldwide in 2026 and highlighted the “Herder” supercomputer built on HPE Cray GX5000 using AMD Instinct GPUs and next-gen EPYC “Venice” CPUs. [10]
Oracle: 50,000 MI450 GPUs starting in calendar Q3 2026
Oracle and AMD announced Oracle Cloud Infrastructure would deploy a publicly available AI supercluster powered by 50,000 AMD Instinct MI450 Series GPUs beginning in calendar Q3 2026, expanding afterward, and that these superclusters will be powered by the Helios rack design (MI450 GPUs, EPYC “Venice” CPUs, and next-gen Pensando networking). [11]
AMD’s own Helios framing: open standards + big performance claims (company projections)
In an AMD blog post, the company positioned Helios as built on Meta’s OCP Open Rack Wide design and described MI450 specs such as up to 432GB of HBM4 per GPU and rack-scale performance targets (noting these are engineering projections). [12]
Investor takeaway: If Nvidia’s strength has been “platform + ecosystem,” AMD is signaling it wants to compete on open rack-scale systems (and open networking standards) rather than only on silicon. The market tends to reward that strategy if—and only if—customers adopt at scale.
Roadmap credibility: MI400 in 2026, and Zen 6 details begin to surface
Reuters: MI400 series + rack product planned for 2026
At AMD’s analyst day, Reuters reported AMD’s next-generation MI400 series is set to launch in 2026, and that AMD plans a complete server rack similar to Nvidia’s GB200 NVL72 approach. [13]
Zen 6 disclosures: not just incremental, with data-center focus
On the CPU side, new technical disclosure is feeding the “data center execution” narrative. Tom’s Hardware reported AMD released an early Zen 6 document for developers indicating Zen 6 is positioned as a deliberately wide, throughput-oriented design, pointing to TSMC 2nm-class manufacturing plans and data center implications (including EPYC “Venice”). [14]
This matters for AMD stock because data-center platforms are sold as CPU + GPU + networking stacks. If AMD can strengthen the CPU side while ramping GPUs, it supports the “share-taking platform” thesis.
The next fundamental checkpoint: earnings, guidance, and what’s included (and not included)
Q3 2025: data center growth and PC rebound
Reuters reported that in the September quarter, AMD’s data center segment revenue rose 22% to $4.3 billion, beating estimates, and noted that AMD said it had received licenses to sell modified versions of its MI300-series chips in China but had not begun sales at that time. Reuters also highlighted client/PC momentum, including client segment sales up 46% to $2.8 billion, helped by AI-PC upgrades and a Windows upgrade cycle. [15]
Q4 2025 outlook: ~$9.6B revenue midpoint; MI308 excluded
In its Q3 2025 results release, AMD guided Q4 2025 revenue to about $9.6 billion ± $300 million and non-GAAP gross margin of about 54.5%, explicitly stating the outlook does not include revenue from MI308 shipments to China. [16]
Next earnings date: early February 2026
Earnings calendars and market trackers currently point to Feb. 3, 2026 as AMD’s next earnings report date (timing subject to confirmation). [17]
Why the “MI308 excluded” line is so important into Dec. 26: it creates a clean framework for traders:
- If China shipments remain limited/delayed → guidance stays the anchor.
- If licensing converts to real shipments sooner → investors may start pricing “guidance upside,” even before it hits reported revenue.
Wall Street forecasts and price targets: broadly bullish, but not uniform
Analyst tracking services differ by methodology and coverage universe, but the broad picture is consistent: the Street is generally constructive on AMD, with price targets clustering in the mid-$200s to low-$300s.
- MarketBeat shows a “Moderate Buy” consensus with an average price target around $277.06 (high $380, low $140) based on its tracked ratings. [18]
- StockAnalysis shows a consensus rating of “Buy” with an average target around $239.97, while also listing a median around $277 and a high target around $345. [19]
Meanwhile, Reuters coverage of AMD’s long-term targets captured the key push-pull in sentiment: analysts welcomed ambitious growth goals, but flagged execution risk and questions about the sustainability of AI infrastructure spending and constraints in supply chains. [20]
How to read that into the Dec. 26 session: Price targets may support the “dip-buying” narrative, but when positioning is crowded, the stock can still swing sharply on any sign that demand, margins, or shipments are not matching the hype.
Technical temperature check: moving averages and momentum readings
If you’re watching AMD from a purely tactical standpoint into the Friday open, Investing.com’s technical summary (timestamped to Dec. 24) shows:
- MA(5): ~215.02
- MA(50): ~209.72
- MA(200): ~217.88
- RSI(14): ~59 (often interpreted as moderate bullish momentum) [21]
Treat these as context, not a prediction—especially in a holiday week, when a single headline can overwhelm technical levels.
Market backdrop into Dec. 26: “Santa rally” seasonality meets holiday liquidity
Year-end sessions can behave differently from “normal” trading days.
- MarketWatch highlighted that Dec. 26 has historically been one of the more consistently positive days for U.S. equities and that this date falls inside the widely watched “Santa Claus rally” window. [22]
- Futures coverage heading out of the Christmas holiday also emphasized the broader market tone and the reality of light year-end trading. [23]
AMD doesn’t trade in a vacuum: if semiconductors and mega-cap tech are bid in thin markets, AMD often amplifies that move; if the tape turns risk-off, AMD can amplify that, too.
Pre-market checklist: what AMD traders and investors should watch before the bell
Here’s the practical “before open” list tailored to AMD’s catalysts:
- China signal clarity
- Alibaba / MI308 order reporting
- Whether additional outlets corroborate the 40,000–50,000 figure or add details on delivery timing and scope. [26]
- Hyperscaler and “AI factory” demand pulse
- The Street is mapping AMD’s 2026–2027 upside to system-level deployments (Helios + MI450/MI400). Watch for any customer procurement headlines that validate—or challenge—that ramp. [27]
- Next earnings expectations
- With the next earnings date expected in early February, revisions to revenue/EPS expectations can become more influential as the calendar turns. [28]
- Holiday-session microstructure
- Nasdaq highlights that extended-hours sessions can bring higher volatility and lower liquidity. That matters if AMD gaps in pre-market on a headline. [29]
Bottom line for Dec. 26: AMD remains a “catalyst-driven” AI stock—China and execution are the near-term swing factors
Going into the Dec. 26 open, AMD’s narrative strength is clear:
- long-duration AI demand plus
- mega-partner validation (OpenAI, Oracle) plus
- rack-scale strategy (Helios) and a visible 2026 roadmap
But the near-term stock reaction may hinge on what’s most uncertain:
- China shipment economics and policy stability (MI308 and licensing) [30]
- Whether “reported orders” become confirmed deployments (Alibaba) [31]
- Whether AMD can deliver on the aggressive growth and platform targets it laid out—goals that even bullish analysts have described as ambitious, with execution risk attached. [32]
This article is for informational purposes only and is not investment advice.
References
1. www.nasdaq.com, 2. stockanalysis.com, 3. www.reuters.com, 4. www.reuters.com, 5. ir.amd.com, 6. technode.com, 7. www.investors.com, 8. www.reuters.com, 9. www.amd.com, 10. www.amd.com, 11. www.oracle.com, 12. www.amd.com, 13. www.reuters.com, 14. www.tomshardware.com, 15. www.reuters.com, 16. ir.amd.com, 17. finance.yahoo.com, 18. www.marketbeat.com, 19. stockanalysis.com, 20. www.reuters.com, 21. www.investing.com, 22. www.marketwatch.com, 23. www.investors.com, 24. www.reuters.com, 25. www.reuters.com, 26. technode.com, 27. www.amd.com, 28. finance.yahoo.com, 29. www.nasdaq.com, 30. www.reuters.com, 31. technode.com, 32. www.reuters.com
