After a sluggish start to 2025, Nvidia (NVDA) stock has come roaring back, reflecting improving market sentiment.
Just a few months ago, the narrative surrounding Nvidia was far more cautious. Earlier in the year, the stock faced mounting pressure as concerns grew over potential demand headwinds for its high-performance GPUs. Notably, the launch of DeepSeek’s cost-effective R1 large language model (LLM) in January raised concerns that Nvidia’s premium-priced chips might lose their appeal. Adding fuel to the fire, new U.S. government restrictions on GPU shipments to China wiped out billions in potential revenue.
Overall, Nvidia’s early 2025 narrative was one of headwinds and caution.
But the tide has turned. In just three months, Nvidia stock has surged 51%, making it the first publicly traded company to briefly surpass a staggering $4 trillion market cap. As macroeconomic fears have eased, so has the selling pressure on tech, including Nvidia. Further, the company’s dominance across the AI ecosystem, including hardware, software, and networking, puts it in a solid position to capitalize on exploding AI infrastructure demand.
Nvidia’s recent financial performance demonstrates the robustness of its fundamentals. In its latest quarter, the company reported $44 billion in revenue, representing a 69% year-over-year increase. The Data Center segment led the charge with $39 billion in sales, up 73%, driven by the accelerating demand. Further, the shift from AI training to inference and the rapid buildout of AI factories around the globe augurs well for growth.
The company’s new Blackwell platform is now the key growth catalyst. In Q1, Blackwell chips contributed nearly 70% of Data Center compute revenue, marking the fastest ramp in Nvidia’s history. With its innovative GB200 NVL systems, Nvidia is delivering AI compute at an unprecedented scale and efficiency, achieving the lowest cost per inference token in the industry.
Nvidia’s resilience shows in other areas as well. Networking has emerged as a powerful second growth engine, generating $5 billion in quarterly revenue, a 64% quarter-over-quarter increase. Technologies like NVLink and Spectrum-X are now essential for AI infrastructure, enabling the high-bandwidth, low-latency performance required for next-generation agentic AI, which aims to surpass generative models by facilitating autonomous reasoning and decision-making.