Nvidia (NASDAQ: NVDA) has produced jaw-dropping returns in recent years, with the stock price up 1,110% since the start of 2023. With that kind of gain, calling Nvidia undervalued seems ludicrous.
Here are three reasons Nvidia stock is still a bargain for investors considering it now despite trading near its all-time high set in late October 2025.
Will AI create the world’s first trillionaire? Our team just released a report on the one little-known company, called an “Indispensable Monopoly” providing the critical technology Nvidia and Intel both need. Continue »
On Feb. 25, Nvidia reported $215.9 billion in fiscal 2026 revenue. For context, Nvidia booked $27 billion in fiscal 2023 revenue — representing an eightfold jump in three years.
The massive increase illustrates how artificial intelligence (AI) has fundamentally changed Nvidia. Data center growth deserves virtually all the credit, as Nvidia earned $193.7 billion in data center revenue in fiscal 2026, up from $15 billion in fiscal 2023.
Despite being a much larger company, Nvidia’s margins are actually higher now than they were a few years ago — a testament to its pricing power and customer willingness to pay a premium price for performance.
In fiscal 2026, Nvidia achieved 71% gross margins, 60.6% operating margins, and 55.6% net profit margins — allowing it to rake in a staggering $120.1 billion in net income.
One of the main arguments against buying Nvidia has been the risk that its margins will decline due to weaker pricing power, lower demand, and competition. But Nvidia continues to prove the doubters wrong, not because it is overcharging for its products and squeezing customers, but because it is delivering massive improvements that justify premium pricing.
In its latest earnings release, Nvidia cited research stating that Blackwell Ultra, which is an upgrade to the initial Blackwell architecture, delivers up to 50 times better performance and 35 times lower costs for agentic AI compared to the Nvidia Hopper platform, which predated Blackwell.
Nvidia’s next platform, called Rubin, uses six different chips that achieve even greater performance improvements and cost reductions through what Nvidia calls “extreme codesign.” This is basically integrating software and hardware for rack-scale data center applications — such as designing Nvidia’s graphics processing units alongside NVLink switches rather than as separate offerings.
