NVIDIA delivered another record-breaking quarter in Q3 FY2026, fueled by unprecedented demand for AI infrastructure and rapid adoption of its Blackwell and GB300 platforms. The earnings call, led by CEO Jensen Huang and CFO Colette Kress, offered not only a detailed view of the quarter’s financial performance but also deep insight into the company’s long‑term strategy, platform transitions, and ecosystem momentum. This rundown highlights the most important numbers, strategic takeaways, and standout quotes from Jensen.
Financial Performance Highlights
NVIDIA posted exceptional results for the quarter ended October 26, 2025. According to the full earnings call transcript from The Motley Fool:
Total revenue reached $57 billion, up 62% year over year
Data center revenue hit a record $51 billion, up 66% year over year
Networking revenue surged 162% to $8.2 billion
Compute segment grew 56% on the strength of GB300
Gaming revenue rose 30% to $4.3 billion
Professional visualization revenue climbed 56% to $760 million
Automotive revenue increased 32% to $592 million
GAAP gross margin: 73.4%; Non‑GAAP: 73.6%
Colette Kress highlighted rising inventory and supply commitments—up 32% and 63% respectively—as NVIDIA prepares for continued scale.
Gross margins: mid‑70% range despite higher input costs
No expected revenue contribution from China data center compute in Q4 due to geopolitical restrictions
Strategic and Platform Takeaways
Surging AI Infrastructure Demand
NVIDIA reiterated that demand continues to far exceed supply. As Jensen put it, the “clouds are sold out”, with every generation of GPU—from Blackwell to Hopper to Ampere—fully utilized. Hyperscalers are rapidly shifting from classical machine learning to generative and agentic AI, driving multiyear capex cycles.
Try Tendrill for free
Want to generate your own public shares? Try Tendrill for free.
Share this article
We're building Tendrill to be the smartest, most accurate agent out there.
Blackwell and GB300 Lead the Product Transition
GB300 accounted for roughly two‑thirds of Blackwell revenue in Q3
Blackwell Ultra and the broader Blackwell family show 5× training speedups vs. Hopper
DeepSeek r1 benchmarks demonstrated 10× better performance per watt and 10× lower cost per token compared to H200
Networking Acceleration
Networking growth was driven by NVLink scale‑up, Quantum X InfiniBand, and Spectrum X Ethernet adoption. These technologies remain central to AI factory buildouts across hyperscalers and enterprises.
Rise of AI Factories
Jensen doubled down on the concept of AI factories—data centers built to generate continuous AI output rather than simply storing or retrieving data. Across customers, roughly 5 million GPUs are tied to new hyperscaler, sovereign AI, and enterprise AI projects.
Key Customer and Ecosystem Initiatives
Among the partnerships cited on the call:
AWS, Google Cloud, Microsoft Azure, Oracle Cloud: expanding Blackwell and CUDA deployments
Anthropic: new strategic partnership aimed at large‑scale model infrastructure
OpenAI: NVIDIA is assisting in building and deploying 10 gigawatts of AI data center capacity
xAI and Humane: collaborating on a 500‑MW AI data center
First Blackwell wafers from U.S. soil in partnership with TSMC (NYSE: TSM)
Rubin Platform for 2026
Rubin remains on schedule for 2026, with first silicon delivered. Jensen emphasized backward compatibility and ecosystem alignment to ensure rapid adoption.
Notable Jensen Huang Quotes
On AI demand and the "bubble" narrative
“There’s been a lot of talk about an AI bubble. From our vantage point, we see something very different.”
On NVIDIA’s full‑stack advantage
“We excel at every phase of AI—from pre-training and post-training to inference. And with our two-decade investment in CUDA acceleration libraries, we are exceptional at simulations, graphics, and classical machine learning as well.”
On the three simultaneous platform shifts
“The world is undergoing three massive platform shifts at once—the transition from CPU to accelerated computing, the shift from classical ML to generative AI, and the rise of agentic AI.”
On building the global AI infrastructure
“What NVIDIA is doing has never been done before. We’ve created a whole new industry called AI factories… factories that generate every single token instead of retrieving pre‑created information.”
On supply chain scale
“NVIDIA’s supply chain basically includes every technology company in the world… and we have been planning for a very big year.”
On the difficulty of competing
“You're not competing against companies anymore; you're competing against teams—and there aren’t many teams in the world who can build systems of this complexity.”
Final Thoughts
NVIDIA’s Q3 FY2026 call underscored the company’s command over the accelerating AI infrastructure cycle, with Blackwell continuing to scale, GB300 driving compute growth, and networking becoming an essential profit engine. Jensen Huang’s commentary made clear that the industry sits at the intersection of multiple platform revolutions—a backdrop NVIDIA appears uniquely positioned to lead.
For investors, analysts, and industry watchers, the transcript offered a clear signal: NVIDIA’s momentum is structural, not cyclical, and the company is architecting the next era of global compute.