NVDA Q1 2025 Earnings Analysis
Dive into $NVDA Nvidia’s Q1 2025 earnings with review of financial performance, key metrics, operating expenses, dilution, customer growth, future outlook
Financial Results:
↗️$44.1B rev (+69.2% YoY, +12% QoQ) beat est by 2.2%
↘️GM (60.5%, -17.8 PPs YoY)🟡
↘️Operating Margin (48.7%, -16.2 PPs YoY)🟡
↗️FCF Margin (59.4%, +2.1 PPs YoY)🟢
↘️Net Margin (42.6%, -14.5 PPs YoY)🟡
↗️EPS $0.81 beat est by 11.0%
Key Metrics
➡️RPO $1.80B (+38.5% YoY)🟡
Segment Revenue
↗️Data Center $39,112M rev (+73.3% YoY)
➡️Gaming $3,763M rev (+42.2% YoY)🟡
➡️Professional Visualization $509M rev (+19.2% YoY)🟡
↗️Auto $567M rev (+72.3% YoY)
➡️OEM & Other $111M rev (+42.3% YoY)🟡
Revenue by Region
➡️United States $20,739M rev (+53.7% YoY, 47.1% of Rev)🟡
↗️China $5,522M rev (+121.7% YoY, 12.5% of Rev)
➡️Taiwan $7,158M rev (+63.7% YoY, 16.2% of Rev)🟡
↗️Other Countries $10,643M rev (+87.2% YoY, 24.2% of Rev)
Operating expenses
↘️SG&A/Revenue 2.4% (-0.6 PPs YoY)
↘️R&D/Revenue 9.1% (-1.4 PPs YoY)
Dilution
↘️SBC/rev 3%, -0.0 PPs QoQ
↘️Basic shares down -0.7% YoY, -0.0 PPs QoQ🟢
↘️Diluted shares down -1.1% YoY, -0.3 PPs QoQ🟢
Guidance
↘️Q2'25 $45.0B guide (+49.8% YoY) missed est by -2.1%🔴
Key points from Nvidia’s First Quarter 2025 Earnings Call:
Financial Performance
NVIDIA reported Q1 FY2026 revenue of $44 billion, a 69% year-over-year increase, exceeding expectations despite a constrained environment. GAAP gross margin was 60.5%, and non-GAAP gross margin was 61%. Excluding the $4.5 billion write-down tied to H20 export controls, non-GAAP gross margin would have reached 71.3%.
Operating expenses increased 7% GAAP and 6% non-GAAP, mainly from compensation and headcount growth. The company returned $14.3 billion to shareholders through buybacks and dividends.
For Q2 FY2026, revenue guidance is $45 billion ±2%, with non-GAAP gross margin projected at 72%, driven by Blackwell's ramp.
Data Centers
Data center revenue reached $39 billion, a 73% year-over-year gain, accounting for nearly 90% of total revenue. Growth was driven by increasing demand for inference workloads tied to reasoning and agentic AI models.
Blackwell GPUs contributed ~70% of compute revenue as the company completed its transition from Hopper. Major hyperscalers, including Microsoft, Google, and Meta, are deploying ~1,000 NBL72 racks per week or 72,000 GPUs weekly.
Sovereign AI deployments expanded to Saudi Arabia, UAE, Taiwan, and Sweden, highlighting global scale.
Export Control Impact
U.S. export restrictions on H20 GPUs halted China shipments from April 9, resulting in a $4.5 billion write-down and $2.5 billion in lost Q1 revenue. Q2 will reflect an additional $8 billion revenue decline.
NVIDIA identified China’s AI market opportunity as $50 billion, now likely inaccessible under current rules. The company has no compliant product available and continues to assess limited options.
Blackwell Ramp
The GB200 NBL72 system redefined inference economics with up to 30x throughput improvement over prior generation. Software optimization has already enhanced performance 1.5x since launch.
GB300 is now sampling and maintains hardware compatibility with GB200. It includes 50% more HBM and delivers a 50% improvement in FP4 dense inference compute.
Blackwell systems are powering models from OpenAI, Perplexity, Deepseeker One, and others. With 70% of compute revenue now coming from Blackwell, this is NVIDIA’s fastest product ramp to date.
Supply Chain Scaling
To meet demand, NVIDIA is expanding manufacturing, including onshoring in Arizona, Texas, and Houston. Each GB200 NVL72 rack weighs nearly 2 tons and includes 1.2 million components, emphasizing operational scale.
Custom Silicon Strategy
NVIDIA launched NVLink Fusion, enabling semi-custom compute units to integrate directly into its fabric. Partners include MediaTek, Norbel, Alchip Technologies, Astera Labs, Pizizzo, and Qualcomm.
The initiative allows bespoke chip development while anchoring customers to NVIDIA’s proprietary ecosystem.
Automotive
Automotive revenue totaled $567 million, up 72% year-over-year but down 1% sequentially. Growth came from autonomous vehicle adoption and NEV demand.
The company launched its platform with Mercedes-Benz, starting with the CLA model. Collaborations with GM, Agility Robotics, XPeng Robotics, and GE Healthcare are accelerating intelligent vehicle and robotic development.
Gaming and AI PCs
Gaming revenue reached a record $3.8 billion, up 48% sequentially and 42% year-over-year. Success stemmed from RTX 5060/5060 Ti launches, priced from $299, and new laptops starting at $1,099.
NVIDIA now supports Microsoft Copilot+, and the GeForce installed base exceeds 100 million AI-capable PCs.
Nintendo's Switch 2 will feature custom RTX GPUs with DLSS. With 150 million units shipped, this is a high-impact design win.
Professional Visualization
Revenue was $509 million, up 19% year-over-year and flat sequentially due to tariff-related delays. NVIDIA expects a rebound in Q2.
New products include DGX Spark (1 PFLOP) and DGX Station (20 PFLOPS), both targeting high-performance desktop AI computing.
Omniverse continues to expand through integrations with SAP and Schneider Electric, and use cases like robotic fleet management with Pegatron, Accenture, and Keyon Group.
Robotics
NVIDIA introduced Isaac Groot N1, the first open foundation model for humanoid robots, enabling reasoning and skill development.
Adopters include Agility Robotics, Boston Dynamics, and XPeng Robotics. In healthcare, GE Healthcare uses Isaac to develop robotic imaging systems.
Simulation capabilities using Isaac and Omniverse are streamlining development and cutting costs across manufacturing.
AI Scaling and Inference
Inference workloads now require 100x to 1,000x more compute than standard chatbots, driven by reasoning AI.
NVIDIA’s Llama NemoTron models offer 20% higher accuracy and 5x faster inference, available via NIMS (NVIDIA Inference Microservices).
Customers such as Capital One, Cisco, Nasdaq, Shell, and Deloitte reported gains in response time and accuracy using NVIDIA’s AI stack.
Networking
Networking revenue surged to $5 billion, up 64% sequentially, reflecting increased AI cluster deployments.
NVLink72 delivers 130TB/s per rack, equivalent to global internet peak bandwidth. SpectrumX Ethernet, with InfiniBand-level enhancements, added Google Cloud and Meta this quarter.
Networking is now annualizing at over $8 billion, supported by NVLink Fusion and BlueField control plane solutions.
Strategic Partnerships
Key partners include Microsoft, Oracle Cloud, Meta, Google Cloud, Capital One, Cisco, and Shell. NVIDIA's stack is now foundational across cloud, enterprise, telecom, and manufacturing.
Sovereign AI collaborations in Saudi Arabia, UAE, Taiwan, and Sweden reflect infrastructure-scale engagements.
U.S. Manufacturing
NVIDIA is reshoring production to the U.S. TSMC is building 6 fabs and 2 packaging facilities in Arizona.
Manufacturing partners Foxconn (Houston) and Wistron (Fort Worth) are building AI supercomputer facilities.
Volume production from U.S. fabs is expected by year-end, supported by long-term supply commitments.
Tariff and Policy
Q1 systems were delayed by tariff-related issues, though normalization is expected in Q2.
The AI diffusion rule was revoked, which is expected to accelerate U.S.-based AI exports and boost platform adoption globally.
Strategic Outlook
NVIDIA reaffirmed guidance for Q2 revenue of $45 billion and gross margin of 72%.
Jensen Huang identified four long-term growth drivers:
Reasoning inference workloads, sovereign AI infrastructure, enterprise AI adoption, and industrial automation.
He emphasized AI as critical infrastructure, comparing its importance to electricity and the Internet, positioning NVIDIA as a foundational platform for the next industrial age.
Management comments on the earnings call.
Product Innovations
Jensen Huang, President and Chief Executive Officer
"Grace Blackwell and NVLink72 is the ideal engine today, the ideal computer thinking machine, if you will, for reasoning AI."
"Compared to Hopper, Grace Blackwell is some forty times higher speed and throughput. This is going to be a huge benefit in driving down the cost while improving the quality of response with excellent quality of service."
Colette Kress, Executive Vice President and Chief Financial Officer
"Sampling of GB300 systems began earlier this month at the major CSPs, and we expect production shipments to commence later this quarter."
Data Centers
Colette Kress, Executive Vice President and Chief Financial Officer
"The Blackwell ramp, the fastest in our company's history, drove a 73% year-on-year increase in data center revenue."
"GB200 and VO racks are now generally available for model builders, enterprises, and sovereign customers to develop and deploy AI."
"On average, major hyperscalers are each deploying nearly 1,000 NBL72 racks, or 72,000 Blackwell GPUs per week."
Blackwell Growth
Jensen Huang, President and Chief Executive Officer
"Grace Blackwell is in full production. We're off to the races."
"Software optimizations have already improved [Blackwell’s] performance by 1.5x in the last month alone. We expect to continue improving the performance of Blackwell through its operational life."
"Our goal from chip to supercomputer, built in America, within a year. Each GB200 MBLink72 rack contains 1.2 million components and weighs nearly two tons. No one has produced supercomputers on this scale."
Automotive
Colette Kress, Executive Vice President and Chief Financial Officer
"We are now in production with our full-stack solution for Mercedes-Benz, starting with the new CLA, hitting roads in the next few months."
"We are partnering with GM to build the next-gen vehicles, factories, and robots using our AI simulation and accelerated computing."
Robotics
Jensen Huang, President and Chief Executive Officer
"We announced Isaac Groot N1, the world’s first open fully customizable foundation model for humanoid robots, enabling generalized reasoning and skill development."
"The era of robotics is here. Billions of robots, hundreds of millions of autonomous vehicles, and hundreds of thousands of robotic factories and warehouses will be developed."
AI Scaling
Jensen Huang, President and Chief Executive Officer
"We’ve reached an extraordinary milestone. With AIs that are reasoning, are thinking, what people call inference time scaling... inference is going to be a significant part of the compute workload."
"Reasoning is compute-intensive and requires hundreds to thousands more tokens per task than previous one-shot inference."
"Inference, once the light of workload, is surging with revenue-generating AI services. AI is growing faster and will be larger than any platform shifts before, including the Internet, mobile, and cloud."
Strategic Partnerships
Colette Kress, Executive Vice President and Chief Financial Officer
"We also announced a partnership with Yum Brands... to bring our AI to 500 of its restaurants this year, expanding to 61,000 restaurants over time."
"Leading platform companies including Accenture, Cadence, Deloitte, and Microsoft are transforming work with our reasoning models."
Jensen Huang, President and Chief Executive Officer
"Every nation now sees AI as core to the next industrial revolution, a new industry that produces intelligence and essential infrastructure for every economy."
US Manufacturing
Jensen Huang, President and Chief Executive Officer
"TSMC is building six fabs and two advanced packaging plants in Arizona to make chips for us. Volume production is expected by year-end."
"In Houston, we're partnering with Foxconn to construct a million-square-foot factory to build AI supercomputers. Wistron is building a similar plant in Fort Worth, Texas."
Challenges and Tariff
Colette Kress, Executive Vice President and Chief Financial Officer
"Tariff-related uncertainty temporarily impacted Q1 systems. Demand for our AI workstations is strong, and we expect sequential revenue growth to resume in Q2."
China Export Controls
Colette Kress, Executive Vice President and Chief Financial Officer
"We recognized $4.6 billion in H20 revenue in Q1... but also recognized a $4.5 billion charge as we wrote down inventory and purchase obligations."
"For Q2, we expect a meaningful decrease in China data center revenue. Our outlook reflects a loss in H20 revenue of approximately $8 billion."
Jensen Huang, President and Chief Executive Officer
"China is one of the world’s largest AI markets and a springboard to global success... Today, however, the $50 billion China market is effectively closed to US industry."
"Shielding Chinese chipmakers from US competition only strengthens them abroad and weakens America's position."
Future Outlook
Jensen Huang, President and Chief Executive Officer
"This is the start of a powerful new wave of growth. Grace Blackwell is in full production."
"The foundations of our next growth platforms are in place and ready to scale — sovereign AI, enterprise AI, industrial AI."
"AI is becoming essential infrastructure, like electricity or the Internet. Every country will have it. Every industry will use it."
Thoughts on Nvidia Earnings Report $NVDA:
🟢 Positive
Revenue rose to $44.1B (+69.2% YoY, +12% QoQ), beating estimates by 2.2%.
EPS of $0.81, beat by 11.0%.
Free Cash Flow margin expanded to 59.4% (+2.1 PPs YoY).
Data Center revenue surged to $39.1B (+73.3% YoY), driven by Blackwell ramp and AI inference demand.
Blackwell GPUs now contribute ~70% of compute revenue; 1,000+ racks/week deployed at hyperscalers.
Gaming revenue hit a record $3.8B (+42.2% YoY, +48% QoQ), boosted by RTX 5060 series.
Auto revenue grew 72.3% YoY to $567M, led by Mercedes-Benz and NEV momentum.
Networking revenue climbed to $5B (+64% QoQ), annualizing over $8B.
Strong partnerships with Microsoft, Meta, Google Cloud, and sovereign nations (Saudi Arabia, Taiwan, Sweden).
Diluted shares down 1.1% YoY, aiding EPS growth.
🟡 Neutral
Professional Visualization revenue was $509M (+19.2% YoY, flat QoQ); impacted by tariff delays.
RPO rose to $1.8B (+38.5% YoY), signaling moderate forward visibility.
SG&A and R&D declined as a % of revenue, but total opex rose with headcount.
U.S. revenue at $20.7B (+53.7% YoY), 47.1% of total—still dominant but slowing relative to other regions.
Custom silicon strategy with NVLink Fusion gaining partners, but monetization still developing.
🔴 Negative
GAAP Gross Margin fell to 60.5% (−17.8 PPs YoY).
Operating Margin dropped to 48.7% (−16.2 PPs YoY).
Net Margin declined to 42.6% (−14.5 PPs YoY).
Q2 Revenue Guide of $45.0B (+49.8% YoY) missed estimates by −2.1%.
$4.5B write-down tied to H20 export restrictions; $2.5B unshipped Q1 revenue and $8B hit expected in Q2.
China AI market (~$50B) likely lost due to regulatory blocks; no compliant product available.
Thank you for reading!
Follow me for more frequent updates on X/Twitter and Threads, and on LinkedIn. For visual infographics, check out Instagram, and for portfolio changes, follow me on Savvy Trader.
Disclaimer: This earnings review is for informational purposes only and does not constitute financial, investment, or trading advice.