Meta deal for Nvidia chips is a big deal. These 2 charts illustrate why
Meta Platforms ‘ pledge to spend billions of dollars on Nvidia chips is a much-needed shot in the arm for the AI semiconductor giant and its recently middling stock. Nvidia’s scorching multiyear rally had cooled off in recent months, as investor dollars moved into other buzzy corners of the chip market like memory and storage, and Google’s impressive AI models built on its in-house chips fanned competition concerns. But now, Meta’s commitment to Nvidia should remind the market of Nvidia’s technology advantages and its central role in the broader AI buildout. “Nvidia has been such a drag on this market that you begin to start a new narrative, which is that we just don’t go after Sandisk and Western Digital and Micron over and over and over again,” Jim Cramer said Wednesday on CNBC. Instead, he said, traders and investors may start to say, “Let’s go back to the ones with great intellectual property.” On that list of IP powerhouses, Nvidia is at the top of the heap. “It’s just a change of scenery” that could be in order, Jim said. Shares of Nvidia rose more than 2% on Wednesday, outperforming the S & P 500 . So far in 2026, however, the stock has gained just over 1%, slightly ahead of the broader market. The two charts below help illustrate just how much the scenery within the semiconductor landscape had changed compared with the early years of the AI boom, when Nvidia’s cutting-edge processors, known as graphic processing units (GPUs), were the hottest commodity in technology, sending its profits and stock price soaring. That “hottest commodity” designation in recent months has shifted to memory chips and data storage devices — technology that’s crucial for AI models being training and used in the real world. But up until recently, it was relatively unsung compared to chips like Nvidia’s, which do the workhorse calculations that underpin AI models. With demand for memory and storage accelerating, a supply shortage has ensued, causing prices for things like DRAM, hard drives and solid state drives (SSDs) to skyrocket. Some investors may also worry that higher memory prices could be passed onto Nvidia’s customers, limiting demand for GPUs if more of the budget goes elsewhere. That is why the charts of Micron, Western Digital, Sandisk, and Seagate look like this since the start of August compared to Nvidia. Nvidia ( blue line at the bottom ) is up merely 4% on a total return basis, trailing by so much it’s hard to see. Sandisk ( pink line ), up more than 1,200%, while Western Digital ( green line ) and Micron ( orange line ) have more than tripled. Seagate ( turquoise line ) is up about 166%. In our minds, some of the money that’s flown into these stocks could’ve been coming from Nvidia. For investors who wanted to ride the memory-and-storage momentum without increasing their exposure to the semiconductor industry, Nvidia would be a convenient a source of funds — especially if you’re sitting on profits on paper. Additionally, Intel ‘s stock (purple line ) saw renewed interest from investors over time time, as the Trump administration took an equity stake, signaling its status as a “national champion.” Even Advanced Micro Devices ( yellow line ), seen as an alternative for customers trying to reduce their reliance on Nvidia, performed ahead of Nvidia during this stretch, as the chart above shows. Another big question market that lingered over Nvidia in recent months was Google’s emergence as an AI darling, fueled in large part by its Gemini 3 model unveiled in mid-November that was trained exclusively on its custom chips, known as Tensor Processing Units (TPUs). Google has co-designed these chips with fellow Club name Broadcom for years, largely for internal use while also offering them via Google Cloud to customers. But the success of Gemini 3 and media reports that Google was in talks to sell TPU servers externally — most prominently to Meta for the social media giant to use in its own data centers — increased the level of concern about Google’s competitive threat to Nvidia. The chart below shows how Nvidia’s performance (blue line) has lagged behind Google parent Alphabet ( orange line ) since Nov. 17 and a broad fund of chip stocks, the iShares Semiconductor ETF , which often referred as its trader symbol SOXX ( pink line ). Curiously, Broadcom ( green line ) has trailed Nvidia during this stretch — but we see that as the market making a mistake, rather than indicative of fundamental weakness in its business. After all, Broadcom helps enable the success of the TPU. Nevertheless, the chart below illustrates the lack of momentum in Nvidia’s shares during this period of hype around Google’s AI prowess, which, it should be noted, is the reason we added it back to our portfolio in late December . Does Meta’s multiyear commitment to buying Nvidia chips mean it will never use TPUs in the future? No, we can’t say that for sure, given Meta and so many other companies developing AI products are hungry for all the computing capacity they can get. There’s also no indication that Meta is abandoning its own custom-chip initiatives, believed to be in partnership with Broadcom, whose CEO, Hock Tan, sits on Meta’s board of directors . But with the market fretting about Nvidia’s competitive standing and the idea that its chips are too expensive so customers must look elsewhere for compute, Meta’s decision to spend billions more on its silicon should not be ignored. That’s especially true considering Meta said it will also deploy Nvidia’s central processing units (CPUs) on a standalone basis — not just in conjunction with its bread-and-butter GPUs, a notable new avenue for Nvidia. Combine that with Meta’s usage of Nvidia’s networking technology, it’s a reminder of just how complete its product portfolio is. Bottom line Meta is lending a real vote of confidence in the “total cost of ownership” of Nvidia’s technology, Jim said Wednesday. He said looking only at the upfront purchase price and not considering the value provided over the lifetime of usage is too narrow. Meta CEO Mark Zuckerberg understands it, Jim said, but “traders don’t think about this at all.” (Jim Cramer’s Charitable Trust is long NVDA, GOOGL, AVGO, and META. See here for a full list of the stocks.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust’s portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
<
