WARNING: If You Hold NVIDIA Stock (NVDA)... GET READY
Quick Overview
NVIDIA stock remains a strong buy despite its recent high valuation because its next-generation AI hardware, particularly the Blackwell NVL72 system, promises 10x performance improvements in inference, drastically lowering the cost per token and supporting an explosion in AI use cases, making it the most valuable company on the analyst's list.
Key Points: NVIDIA reported Q3 FY26 revenue of $57.0B, up 22% sequentially, with Data Center revenue hitting $51.22B (88% of total revenue), showing 25% quarter-over-quarter growth. Blackwell NVL72 systems offer 10x performance improvement over Hopper, achieving 10x lower cost per token and supporting 3x to 6x more users for the same power budget. The Blackwell/Rubin revenue visibility extends to over half a trillion dollars cumulatively through the end of CY2026, exceeding Hopper's lifetime revenue by more than 5x. NVIDIA's Data Center networking revenue grew to $8.2B, exceeding gaming revenue (9% of total) and establishing NVIDIA as the world's largest networking business. NVIDIA's forward Price-to-Earnings (PE) ratio of 28.7x is significantly lower than peers like AMD (56.2x) and Broadcom (64.5x), despite NVIDIA being the most valuable company. The analyst strongly recommends buying NVDA stock now, predicting it will become the world's first $10 trillion company within the next couple of years. The video features an interview with Dion Harris, NVIDIA's Sr. Director of HPC, Cloud, and AI Infrastructure GTM, discussing these performance and cost metrics.
Context: This video analyzes NVIDIA's latest earnings report, focusing on the exponential growth driven by its Data Center business and the disruptive potential of its upcoming Blackwell architecture, specifically the NVL72 system. The presenter contrasts current performance metrics (Hopper) with projected performance (Blackwell) to argue for continued stock appreciation, despite high market expectations. The analysis also touches upon broader AI market growth and how next-generation hardware efficiency translates directly into lower operational costs for data centers.