Meta shares pulled back after a strong rally in the prior session, easing off even as Wells Fargo raised its price target on the stock to $767. The intense focus on the company followed reports that it is developing a program dubbed "Meta Compute." The initiative reportedly aims to generate revenue from excess AI computing capacity by selling raw compute to other businesses. If executed, this would place Meta in direct competition with dedicated compute providers such as CoreWeave and the neocloud provider Nebius, as well as with the established hyperscalers.
The Bull Case: A Brilliant Way to Monetize Capex
There is a compelling argument that this is a shrewd strategic move by Mark Zuckerberg and his team. For a long time, investors have been demanding to see a return on the enormous invested capital that Meta has poured into AI. The company has spent hundreds of billions of dollars over the past couple of years building out compute and the broader AI infrastructure it needs. Meta's reasoning is straightforward: if it has overcapacity in compute, it can simply sell the surplus rather than let it sit idle.
This announcement had a notable ripple effect. It put a crimp in the memory chip stocks, a drag that began the previous day and carried forward. While it may be overstating things to call it a "death knell" for that trade, it was at least enough to make some investors reconsider their positioning in memory names.
The initial market reaction to the news was fairly strong. The stock rallied nicely in the prior session and opened as though it might continue higher, but the momentum began to fade roughly an hour into trading.
Why Diversification Matters
The strategic logic rests on Meta's revenue concentration. The company relies on advertising for over 95% of its revenue, which creates a clear need to diversify. Unlike Alphabet (Google) or Amazon, Meta does not have an established cloud business — an important piece of context. What Meta does possess is scale in attention: more than 3.6 billion daily "eyeballs" across its platforms globally. That massive audience is precisely why its advertising business performs so well.
The open question is one of monetization efficiency. People are engaged on these social media platforms, but it is unclear whether Meta can monetize that engagement as effectively as the hyperscalers and cloud companies monetize compute. Selling compute is likely part of the answer to that diversification problem.
The economics are striking. Selling just one gigawatt of energy within AI and compute equates to roughly $20 billion in revenue annually. If Meta genuinely has the spare capacity, the move makes obvious sense — prompting the wry observation that investors and analysts may be "in the wrong business" and should be selling gigawatts themselves, since that is where the money is.
The Bear Case and the Lingering Questions
Despite the appeal of the strategy, meaningful questions remain. Meta is simply not as diversified a business as some of the other big tech names, and that is a persistent concern even if the compute-selling initiative succeeds.
The stock's price action reinforces a cautious read. Recent trading has followed a pattern where good news produces a small pop, but the stock cannot sustain it — it continues to grind within an intermediate range. It "wants" to move higher but repeatedly fails to reclaim its high point around 650, and then fades back toward its low point around 550.
This 100-point band may look wide, but it is best understood in terms of ratios. A move between a $55 stock and a $65 stock would not seem like a wide range, and 550 to 650 is proportionally the same thing.
On the technicals, the stock failed at its 50-day moving average, located around 605, on the recent push higher, and has since been heading back down, trading around 585. There is reason to expect it could drift toward its 21-day exponential moving average near 581 — not far below current levels. Notably, 540 was the low tick hit on June 26, making that area a plausible zone where buyers might step back in. The practical takeaway from this bearish-leaning technical read is to fade any pop that appears in the stock, an objective that can be accomplished with a short upside call vertical.
Example Trade One: A Short Call Vertical
The first strategy is designed to fade the rally and build on the intraday fade. It uses the July 17th monthly expiration — roughly two weeks, or 15 days, to expiration — keeping the position short-term and taking advantage of elevated implied volatility.
Implied volatility on Meta sits around the 66th percentile, which is relatively high for the name, and it is likely to keep building as the stock approaches earnings and then expiration. That elevated volatility, combined with the drop below the 50-day moving average, helped construct the trade.
The specific structure is a short neutral-to-bearish call vertical: sell the July 17th 610 call and buy the 615 call, a spread that is $5 wide. At the time of the trade, it could be sold for a credit of about $1.35, which is the maximum profit. Because the spread is $5 wide, the risk is $365. Collecting the $1.35 credit pushes the break-even all the way up to 611.35, providing a large cushion to the upside.
The probabilities favor the seller. There is roughly a 71% probability that the short 610 strike finishes out of the money at expiration — which is the goal. If the stock stays below 611.35, the trade keeps the collected credit (or the position can be bought back cheaper while the market is open). This is a passive directional trade that profits if the stock goes lower, stays where it is, or even rises somewhat, as long as it remains below the break-even. The trade-off inherent in this high-probability strategy is that you risk more to make less.
Example Trade Two: An Unbalanced Put Butterfly for a Credit
The second approach keeps the passive character but changes the structure to an unbalanced put butterfly that also collects a credit. It goes out to the July 24th weekly option, giving an extra week of time. Since Meta reports earnings on the 29th, this still avoids the event risk of earnings.
The structure buys one 575-strike put, sells two 570-strike puts, and buys one 540-strike put — an unbalanced put butterfly in the July 24th weekly cycle. It was trading for a credit of about $5.60 earlier, and slightly more might be collected. That $5.60 credit is the maximum profit if the stock stays above 575, above the strike that was bought. Anything above that level and the trade simply keeps the credit, making it a neutral-to-bullish position.
The distinctive feature is the "kicker." As the stock slides toward the 570 strike — where two options were sold — profitability spikes to over $1,000 near that short strike, which sits about 16 points, or roughly $16, out of the money.
The risk lies to the downside. The break-even is just below 559.50; the position needs to remain above that level. Below 540, the trade risks about $1,930. That 540 level is meaningful because it was the June 26 low tick, a decent area where buyers might re-emerge. Like the call vertical, this structure profits in three out of four scenarios, with the added kicker if the stock drifts down to that 570 level.
Overall, the put butterfly is a bullish trade, but its structure still provides some cushion to the downside should a pullback occur — an interesting way to express a constructive view while retaining a margin of safety.
