Loading…
Loading…
$202.81
Market Capi
$5.0T
Growth-adj P/E (3-yr hist)i
0.2x
Growth-adj P/E (3-yr proj)i
0.3x
P/S 27Ei
13x
P/S 29Ei
7.3x
EV/EBIT 27Ei
19x
EV/EBIT 29Ei
9.7x
P/E 27Ei
22x
P/E 29Ei
13x
P/S 27Ei
13x
EV/EBIT 27Ei
19x
P/E 27Ei
22x
P/S 29Ei
7.3x
EV/EBIT 29Ei
9.7x
P/E 29Ei
13x
Revenue
2027E
$372B
Gross Margini
71%
Hist. CAGRi
100%
Proj. CAGRi
40%
EBIT
2027E
$242B
Op. Margini
60%
Hist. CAGRi
186%
Proj. CAGRi
44%
Net profit
2027E
$202B
Net Margini
56%
Hist. CAGRi
202%
Proj. CAGRi
37%
Business Model
Recent Developments
Average Targeti
$301.62+48%
Consensusi
Buy
61 analysts covering
Net cashi
$160B
Div. Yieldi
0.4%
Buyback Yldi
0.8%
Market Capi
$5.0T
P/E 27Ei
22x
Adj. P/E (fwd.)i
0.3x
Revenue 27E
$372B
Proj. CAGRi
40%
Gross Margini
71%
EBIT 27E
$242B
Proj. CAGRi
44%
Op. Margini
60%
Net Profit 27E
$202B
Proj. CAGRi
37%
Net Margini
56%
Average Targeti
$301.62+48%
Consensusi
Buy
61 analysts covering
Profile
NVIDIA designs and sells accelerated computing platforms — GPUs, networking silicon, and the software stack that ties them together — primarily for AI training and inference workloads. The company does not manufacture its own chips; it fabless-designs them and contracts production to TSMC.
Industry
NVIDIA's primary customers are hyperscalers and cloud providers — Microsoft, Google, Amazon, Meta, and Oracle collectively account for a large share of Data Center revenue. Beyond the largest cloud operators, the customer base extends to AI-native companies, sovereign AI programs, and HPC research institutions. Geographically, the United States contributes roughly 47% of revenue, with Singapore (18%), Taiwan (16%), and China/Hong Kong (13%) rounding out the mix. The market is oligopolistic at the high end: switching costs are steep given deep CUDA dependency, and lead times on next-generation GPUs are often measured in years rather than quarters.
Key metrics
Economic moat
CUDA is the primary moat: 25+ years of developer tooling, libraries (cuDNN, TensorRT, RAPIDS), and hardware-software co-design have created a switching cost that AMD and Intel have not been able to erode despite years of effort. Beyond software, scale in chip architecture R&D means NVIDIA can invest more per node transition than any competitor. In networking, the Mellanox acquisition (2020) gave NVIDIA end-to-end control of AI fabric — a structural advantage as cluster sizes scale beyond single-rack boundaries. Brand among AI researchers functions as a pull channel: the default assumption for any new model training run is NVIDIA hardware.
Jensen Huang
Huang co-founded NVIDIA in 1993 alongside Chris Malachowsky and Curtis Priem, and has served as CEO continuously since. He holds a BS in electrical engineering from Oregon State and an MSEE from Stanford, and spent formative years as a chip designer at AMD and LSI Logic before starting the company. His decision to pivot NVIDIA from gaming GPUs toward general-purpose parallel computing — and the 2006 launch of CUDA — proved to be the most consequential product strategy in semiconductor history. Huang remains the dominant strategic voice at NVIDIA and owns a meaningful equity stake, aligning his interests with long-term shareholders.
Colette Kress
Kress joined NVIDIA as CFO in September 2013, bringing experience from Cisco (SVP Finance) and Microsoft. She has overseen NVIDIA through a roughly 100x increase in revenue, managing capital allocation decisions including the $6.9 billion Mellanox acquisition in 2020. Her consistency in managing analyst expectations and guidance discipline is a key element of NVIDIA's credibility with institutional investors.
Debora Shoquist
Shoquist has served as Chief Operating Officer since September 2007, with responsibility for operations, supply chain, and global manufacturing partnerships. Her tenure spans NVIDIA's entire transition from a consumer GPU company to the dominant AI infrastructure supplier, including the critical management of TSMC supply during periods of extreme demand. Supply chain execution — particularly GPU allocation during the Blackwell ramp — is among the most operationally complex challenges in the semiconductor industry, and her continuity in the role is a stability factor.
Other key figures
Marco Pavone (joined January 2021) leads robotics and autonomous systems research, reflecting the company's push into physical AI and the DRIVE platform. Toshiya Hari took over as Investor Relations Contact in February 2025. The board includes long-tenured members Harvey Jones and Tench Coxe (both directors since 1993), providing institutional continuity alongside newer independent additions.
P/S Ratioi
EV/EBITi
P/E Ratioi
Revenue
CAGR (hist. 3-yr)i
0%
CAGR (proj. 3-yr)i
0%
EBIT
CAGR (hist. 3-yr)i
0%
CAGR (proj. 3-yr)i
0%
Net profit
CAGR (hist. 3-yr)i
0%
CAGR (proj. 3-yr)i
0%
Values in billions of USD.
Operating cashflow · Levered Free Cash Flow
Free Cash Flow
CAPEX
Values in billions of USD.
Margins
Rentability
Balance sheet
Values in billions of USD.
Liquidity ratios
Debt-to-Equity-Ratio
Last earnings
Revenue of $44.1 billion for Q1 FY2026 (ended April 27, 2025) grew 69% year-over-year and 12% sequentially, broadly in line with consensus.
Non-GAAP EPS came in at $0.81, or $0.96 excluding a one-time $4.5 billion charge tied to H20 GPU inventory written off after the U.S. government imposed new export licensing requirements on China sales in April 2025.
Data center revenue hit $39.1 billion, up 73% year-over-year, driven by rapid adoption of Blackwell-based NVLink and Ethernet AI solutions. Networking revenue within that segment surged 64% sequentially.
Q2 FY2026 guidance was set at $45 billion (±2%), explicitly absorbing roughly $8 billion in lost H20 sales to China. CEO Jensen Huang noted that AI inference token generation had surged tenfold in a single year.
Recent developments
On May 6, 2026, NVIDIA and Corning announced a long-term manufacturing partnership: Corning will expand U.S. optical connectivity capacity tenfold and fiber production by over 50%, funded in part by an NVIDIA right to invest up to $3.2 billion.
On May 7, 2026, NVIDIA and data center operator IREN announced a strategic partnership to deploy up to 5 gigawatts of NVIDIA DSX-aligned AI infrastructure globally, with NVIDIA receiving a five-year option to invest up to $2.1 billion in IREN shares.
At CES 2026 in January, NVIDIA launched the Vera Rubin platform — six co-designed chips including the Rubin GPU and Vera CPU. Rubin entered full production and Rubin-based products are expected from major cloud providers in the second half of 2026, promising a 4x reduction in GPUs needed for mixture-of-experts training versus Blackwell.
NVIDIA is scheduled to report Q1 FY2027 earnings on May 20, 2026, with consensus expecting approximately $78.8 billion in revenue — a 78% year-over-year increase — against company guidance of $78 billion.
Debate & sentiment
Bulls point to structural AI infrastructure demand: with 60 analysts covering the stock and consensus at Buy, the investment thesis rests on hyperscaler capex continuing to accelerate and Blackwell supply transitioning smoothly to Rubin in H2 2026.
The China question is the central bear risk: export controls forced a $4.5 billion write-down in Q1 FY2026 and eliminated roughly $8 billion in Q2 revenue. Domestic Chinese alternatives from Huawei and others are gaining share, with TrendForce projecting local chips at roughly 50% of China AI shipments by end of 2026.
A potential China re-opening is a latent upside catalyst: NVIDIA guidance already assumes zero H20 and H200 revenue from China, so any regulatory softening — reports indicate H200 shipments are being prepared under a new framework — would force analyst estimates upward from a zero baseline.
Valuation has re-rated sharply: with a forward PEG of 0.9 and FCF yield near 5.75%, the stock is no longer priced for perfection on 2026 estimates, but execution on Rubin ramp and margin recovery above the H20-impacted 61% gross margin are the key variables to watch.
Consensusi
Buy
Average targeti
$301.62+48%
Highest targeti
$500.00+146%
Lowest targeti
$180.00-12%
Nice, you made it all the way through!
Free for 14 days, then $9/month for every report, every metric, and every feature. Straightforward, as it should be. Future additions included. Cancel anytime.