Research Finder
Find by Keyword
NVIDIA Reports Robust Q2 2026 Financial Results, Shines Spotlight on Blackwell
NVIDIA delivered strong Q2 2026 results that surpassed both its own guidance and Wall Street's expectations, although continued exclusion of H20 sales from its future outlook highlights persistent uncertainties regarding its business in China.
Key Highlights
- NVIDIA reported a Q2 revenue of $46.7 billion, marking a substantial 56% increase from the previous year, driven primarily by its data center segment.
- The company's Blackwell Data Center revenue grew 17% sequentially, with new hardware like the RTX PRO 6000 Blackwell Server Edition GPU and Spectrum-XGS Ethernet being adopted by major companies worldwide.
- NVIDIA's Board of Directors authorized an additional $60 billion for its share repurchase program, signaling confidence and a commitment to returning value to shareholders.
- The company had no sales of its H20 chips to China in Q2, and the exclusion of these sales from its Q3 guidance reflects ongoing geopolitical uncertainties and a structural shift toward AI self-sufficiency in the Chinese market.
- NVIDIA is expanding its reach beyond data centers, with strong growth in its Gaming and Automotive/Robotics segments, and is leveraging its software and hardware ecosystem to fuel new markets like cloud gaming and AI-powered robotics.
The News
NVIDIA (NASDAQ: NVDA) reported revenue for the second quarter (Q2) ended July 27, 2025, of $46.7 billion, up 6% from the previous quarter and up 56% from a year ago. NVIDIA’s Blackwell Data Center revenue grew 17% sequentially. For more information, read the NVIDIA press release.
Analyst Take
NVIDIA announced a very strong fiscal Q2, with revenue reaching $46.7 billion. This figure represents a 6% increase from the previous quarter and a remarkable 56% jump compared to the same period last year. The primary driver of this growth was the company's Blackwell Data Center revenue, which saw a significant 17% sequential increase.
The report also detailed key aspects of the company's product sales. During the quarter, NVIDIA had no sales of its H20 products to customers in China. However, the company's revenue was boosted by approximately $650 million in unrestricted H20 sales to a customer outside of China, which included a $180 million release of inventory that had been previously reserved.
NVIDIA reported both its GAAP and non-GAAP gross margins at 72.4% and 72.7%, respectively. The company's non-GAAP gross margin would have been slightly lower, at 72.3%, without the $180 million release of previously reserved inventory. Similarly, the company's GAAP earnings per diluted share stood at $1.08, while non-GAAP earnings were $1.05. Excluding the financial benefit from the inventory release and its related tax impact, the non-GAAP diluted earnings per share would have been $1.04.
NVIDIA's commitment to returning value to its shareholders was clear in the first half of fiscal 2026, as the company returned a total of $24.3 billion through share repurchases and cash dividends. As of the end of the second quarter, NVIDIA still had $14.7 billion remaining in its existing share repurchase program.
To further emphasize this strategy, the Board of Directors approved an additional $60.0 billion for its share repurchase authorization on August 26, 2025. This new authorization has no expiration date, giving the company significant flexibility to continue its buyback program.
Data Center Takeaways
For Q2, NVIDIA reported revenue of $41.1 billion, representing a 5% increase from the previous quarter and a significant 56% increase compared to a year ago. The company also announced that the new NVIDIA RTX PRO 6000 Blackwell Server Edition GPU is being adopted by some of the world's most popular enterprise servers, with companies such as Disney, Foxconn, Hitachi, Hyundai, Lilly, SAP, and TSMC being among the first to use them. Additionally, NVIDIA introduced NVIDIA Spectrum-XGS Ethernet to connect distributed data centers for giga-scale AI.
NVIDIA is working with several European nations, including France, Germany, Italy, Spain, and the U.K., as well as key technology leaders to build NVIDIA Blackwell AI infrastructure. This initiative includes the creation of the world's first industrial AI cloud for European manufacturers to drive the region's next industrial transformation. The company also announced the expansion of NVIDIA DGX Cloud Lepton to connect Europe's developers to its global compute ecosystem.
NVIDIA is collaborating with global partners to accelerate advanced AI supercomputers like Doudna in the U.S., JUPITER and Blue Lion in Germany, Isambard in the U.K., and FugakuNEXT in Japan. Furthermore, the company revealed that model builders in Europe and the Middle East are optimizing their sovereign large language models with NVIDIA Nemotron, which will be available on Perplexity.
In the quarter, NVIDIA supported the launch of OpenAI's open gpt-oss models, delivering a performance of 1.5 million tokens per second on a single NVIDIA Blackwell GB200 NVL72 rack-scale system. The company also announced a collaboration with Novo Nordisk and DCAI to advance drug discovery. The NVIDIA Blackwell platform demonstrated its top performance by delivering the highest results at scale on every MLPerf Training benchmark. NVIDIA also teamed up with Ansys and DCAI to advance quantum algorithms for fluid dynamics using the NVIDIA CUDA-Q platform. Additionally, the company introduced NVFP4, a 4-bit format designed to provide exceptional inference latency for pretraining next-generation large language models.
Data Center Drill Down: Spectrum-XGS Ethernet Unveiled
As the demand for AI grows, individual data centers are hitting their limits in terms of power and capacity. To keep up, data centers need to expand beyond a single location, but traditional Ethernet networks can struggle with this because of high latency and inconsistent performance. NVIDIA unveiled NVIDIA Spectrum-XGS Ethernet, a new technology designed to overcome these challenges.
It is an addition to the NVIDIA Spectrum-X Ethernet platform that allows multiple data centers to be combined into a single, massive AI super-factory. Think of it as a third way to scale AI computing, beyond just scaling up within a single server or scaling out within a single data center. Spectrum-XGS makes it possible to connect distributed data centers, creating a unified, giga-scale system with predictable performance.
Spectrum-XGS Ethernet is a new feature within the Spectrum-X platform that's designed to connect distant data centers. It uses advanced algorithms that automatically adjust the network to account for the distance between locations. By managing congestion and latency, and providing full network monitoring, it almost doubles the performance of the NVIDIA Collective Communications Library (NCCL). This means communication between multiple GPUs and servers is faster and more reliable, allowing multiple data centers to function as one cohesive AI super-factory.
The NVIDIA Spectrum-X Ethernet platform provides 1.6x greater bandwidth than off-the-shelf Ethernet, which I expect can make it the top choice for massive, multi-tenant AI supercomputers. Featuring NVIDIA Spectrum-X switches and ConnectX-8 SuperNICs, the platform delivers the low latency and high performance essential for building scalable AI.
From our perspective, NVIDIA Spectrum-XGS Ethernet's ability to deliver competitive advantages are primarily centered on its scale-across breakthrough technology, which is specifically optimized to turn multiple, geographically dispersed data centers into a single, cohesive AI super-factory. As such, we anticipate that large-scale AI factories are the new essential infrastructure. With NVIDIA Spectrum-XGS Ethernet, NVIDIA is adding the new scale across dimensions to AI scaling. This technology links data centers across cities, nations, and continents, creating a single, massive AI super-factory.
AI PC and Gaming Drill Down: Making AI More Accessible
In Q2, NVIDIA's Gaming revenue was a strong $4.3 billion, showing a 14% increase from the prior quarter and a substantial 49% increase compared to the same period last year. This growth was driven by the launch of the NVIDIA GeForce RTX 5060, a Blackwell-powered GPU that has become the fastest-ramping "x60-class" GPU in the company's history. The quarter also saw NVIDIA DLSS 4 technology integrated into over 175 games and apps, with major upcoming titles such as Borderlands 4, Resident Evil Requiem, and Phantom Blade Zero set to include the feature.
NVIDIA's gaming ecosystem continued to expand, with the announcement that the Blackwell architecture is coming to the cloud gaming service GeForce NOW. This update will also introduce a new "Install-to-Play" feature, which is expected to double the service's game library to over 4,500 titles. Additionally, NVIDIA partnered with OpenAI to launch new open-weight models that are optimized for RTX GPUs, enabling faster, local AI inference in popular tools like Ollama, llama.cpp, and Microsoft AI Foundry Local.
From our perspective, NVIDIA should leverage its comprehensive ecosystem of hardware and software to improve its competitiveness by continuing to innovate with its GeForce RTX GPUs, ensuring the company can offer unmatched performance for both gaming and local AI tasks. This includes accelerating the development of new GPU architectures, such as the Blackwell-based RTX 5060, to maintain a significant performance lead over key rivals AMD and Intel.
Concurrently, NVIDIA should expand its software offerings, such as NVIDIA DLSS, RTX AI, and GeForce NOW, to create a compelling and sticky platform for consumers. This involves making AI-powered features more accessible and useful in everyday PC applications and games, while also expanding its cloud gaming services to attract a wider user base who may not have high-end hardware. By combining competitively advantageous hardware with a powerful and user-friendly software ecosystem, NVIDIA can further solidify its market position and fend off growing competition.
Robotics and Automotive Drill Down: AI Supercomputers Rising
NVIDIA's Automotive and Robotics divisions showed strong growth in the second quarter. The Automotive segment's revenue reached $586 million, marking a 3% increase from the previous quarter and a significant 69% jump from the same period last year. This growth was bolstered by several key announcements, including the full-scale production of the NVIDIA DRIVE AV software platform, which is designed to accelerate the deployment of intelligent transportation. The company also secured its second consecutive win in the "End-to-End Driving at Scale" category of the Autonomous Grand Challenge and began initial shipments of its NVIDIA DRIVE AGX Thor system-on-a-chip.
In robotics, NVIDIA made several key releases to power the next generation of AI-driven machines. It announced the general availability of its NVIDIA Jetson AGX Thor developer kit and production modules, which are powerful new AI supercomputers designed for millions of robots. Additionally, the company released the NVIDIA Halos full-stack safety platform for robotic development and introduced the new NVIDIA Cosmos world foundation models to accelerate the creation and deployment of robotics solutions.
At Hot Chips 2025, NVIDIA unveiled Spectrum-XGS Ethernet as a new feature within the Spectrum-X platform that's designed to connect distant data centers. It uses advanced algorithms that automatically adjust the network to account for the distance between locations. By managing congestion and latency, and providing full network monitoring, it almost doubles the performance of the NVIDIA Collective Communications Library. This means communication between multiple GPUs and servers is faster and more reliable, allowing multiple data centers to function as one cohesive AI super-factory.
The NVIDIA Spectrum-X Ethernet platform provides 1.6x greater bandwidth than off-the-shelf Ethernet, which we anticipate can make it the top choice for massive, multi-tenant AI supercomputers. Featuring NVIDIA Spectrum-X switches and ConnectX-8 SuperNICs, the platform delivers the low latency and high performance essential for building scalable AI.
Professional Visualization Drill Down: Omniverse Ecosystem Collaboration Advances
NVIDIA's Professional Visualization segment showed strong growth in fiscal Q2 2026, with revenue reaching $601 million. This represents an 18% increase from the previous quarter and a substantial 32% jump compared to the same period last year. The company's performance was bolstered by the introduction of new hardware, including the NVIDIA RTX PRO 4000 SFF Edition and RTX PRO 2000 Blackwell GPUs.
Additionally, NVIDIA expanded its collaboration with Siemens to help digitalize and advance the manufacturing factory of the future. The company also announced new NVIDIA Omniverse libraries and software development kits, which are designed to accelerate the development of physical AI solutions.
To strengthen its Professional Visualization segment, we find that NVIDIA should focus on both its hardware and software ecosystems over the next two to four quarters. On the hardware front, it must continue to innovate with its RTX PRO GPUs, ensuring they deliver superior performance, especially for demanding AI-powered creative and design workloads. This includes not just raw power but also optimizing for energy efficiency and form factors, such as the new SFF Edition, to meet the needs of diverse professional workstations and servers. Expanding strategic partnerships with major enterprise companies and independent software vendors (ISVs) will be crucial for accelerating the adoption of new GPUs and ensuring they are certified to work fluently with professional applications.
Looking Ahead
We find it encouraging that NVIDIA's official outlook for the third quarter (Q3) of fiscal 2026 points to another period of robust growth, with the company's data center business as the main driver. The company has set its Q3 revenue forecast at approximately $54.0 billion, with a potential range of plus or minus 2%. It is important to note that this strong guidance does not account for any sales of its H20 chip to China, due to existing export restrictions.
Looking at other key financial metrics, NVIDIA anticipates a non-GAAP gross margin of around 73.5% for the quarter, with expectations to maintain gross margins in the mid-70% range for the full fiscal year. Non-GAAP operating expenses are projected to be about $4.2 billion. Regarding sales to China, while NVIDIA's guidance does not include H20 sales, the company has mentioned that it has secured licenses to restart sales to certain customers in the region. This creates an interesting potential for upside if these sales materialize, although geopolitical uncertainties have led NVIDIA to be cautious and exclude them from its official forecast.
Of concern, NVIDIA's long-term prospects in the crucial Chinese market are facing a structural challenge as China shifts its focus toward achieving self-sufficiency in AI technology. We see several factors are strengthening China's domestic AI chip industry, making it a more formidable competitor to foreign companies like NVIDIA. The explosive growth of local chipmaker Cambricon and the rapid expansion of China's own chip manufacturing capabilities are key drivers. This is further supported by a growing trend of Chinese tech giants complying with the government's push for AI independence, prioritizing homegrown solutions over foreign ones.
This move could have an enduring impact on NVIDIA's business, rather than a temporary one. The company's involvement in the ongoing U.S.-China geopolitical tensions is now seen as a structural setback, not merely a short-term disruption. This situation could reduce NVIDIA’s total addressable market and impact the growth trajectory of its China-related and, as a result, overall revenue stream.
Ron Westfall | Analyst In Residence
Ron Westfall is a prominent analyst figure in technology and business transformation. Recognized as a Top 20 Analyst by AR Insights and a Tech Target contributor, his insights are featured in major media such as CNBC, Schwab Network, and NMG Media.
His expertise covers transformative fields such as Hybrid Cloud, AI Networking, Security Infrastructure, Edge Cloud Computing, Wireline/Wireless Connectivity, and 5G-IoT. Ron bridges the gap between C-suite strategic goals and the practical needs of end users and partners, driving technology ROI for leading organizations.
Share
Stephen Sopko
Analyst-in-Residence – Semiconductors & Deep Tech
Stephen Sopko is an Analyst-in-Residence specializing in semiconductors and the deep technologies powering today’s innovation ecosystem. With decades of executive experience spanning Fortune 100, government, and startups, he provides actionable insights by connecting market trends and cutting-edge technologies to business outcomes.
Stephen’s expertise in analyzing the entire buyer’s journey, from technology acquisition to implementation, was refined during his tenure as co-founder and COO of Palisade Compliance, where he helped Fortune 500 clients optimize technology investments. His ability to identify opportunities at the intersection of semiconductors, emerging technologies, and enterprise needs makes him a sought-after advisor to stakeholders navigating complex decisions.