Back to News
Market Impact: 0.7

NVIDIA Soars 39% in 2025: Will the Stock Carry Momentum in 2026?

NVDAQCOMSTMTXNHIMS
Artificial IntelligenceTechnology & InnovationCorporate EarningsCorporate Guidance & OutlookCompany FundamentalsCapital Returns (Dividends / Buybacks)Analyst EstimatesMarket Technicals & Flows
NVIDIA Soars 39% in 2025: Will the Stock Carry Momentum in 2026?

NVIDIA reported blockbuster fiscal Q3 2026 results driven by its Data Center segment, which generated $51.22 billion (89.8% of sales), up 66% year-over-year and 25% sequentially on strong Blackwell/Hopper GPU demand; overall revenues rose 62% YoY and non-GAAP EPS climbed 60%. The company provided upbeat guidance with Q4 revenue expected at $65 billion (+66% YoY) and a 75% gross margin (up 150 bps YoY), while free cash flow totaled $23.75 billion in the quarter ($66.53 billion YTD), cash balances reached $60.6 billion, and the firm returned $12.46 billion via buybacks in the quarter ($36.27 billion YTD) plus $243 million in dividends. NVDA's forward 12-month P/E of 26.8x trades below the industry average and the stock has outperformed peers year-to-date, supporting Zacks' Buy stance and making the company a high-conviction AI/capital-allocation play for investors.

Analysis

Market structure: NVDA is the clear winner as hyperscalers and enterprises front-load GPU capacity for LLM training/inference, increasing NVIDIA's pricing power and allowing gross margins to target ~75% (per guidance). Direct beneficiaries include cloud providers and ISVs that monetize LLMs; losers are cyclic/analog names (TXN) and legacy mobile suppliers (QCOM) who lack exposure to generative-AI acceleration. Supply/demand now tilts tight for high-end GPUs — demand growth likely >50% YoY for next 4–8 quarters, implying continued outsize revenue share for NVIDIA. Risk assessment: Key tail risks are US/ally export controls to China, protracted TSMC capacity constraints, and a sharp pullback in hyperscaler capex (a >20% YoY spend cut would materially hit guidance). Immediate (days) risk: earnings-driven IV spikes and mean reversion; short-term (0–6 months): guidance misses or inventory destocking; long-term (1–3 years): customer verticals building custom silicon (20–30% share risk). Monitor gross margin inflection points — a drop below 70% would be a red flag. Trade implications: Tactical: establish a core long NVDA allocation via 12–18 month LEAPS (buy Jan 2027 calls ~30% OTM) sized 2–3% NAV and hedge with 6–9 month puts 15–20% OTM if downside protection desired. Pair trade: long NVDA (2%) / short TXN (1.25%) to isolate AI exposure vs analog cyclical risk; target pair profit 20% relative, stop-loss if NVDA underperforms STM/QCOM by >10% in 30 days. Rotate 3–6% cash from broad semiconductor longs into AI infrastructure and datacenter capex names. Contrarian angles: Consensus understates single-customer concentration (hyperscalers ≫50% DC rev) and supply-chain concentration (TSMC). The market may be underpricing regulatory fragmentation — if meaningful export bans return, NVDA multiple could compress from ~27x to <20x. Historical parallel: rapid adoption cycles (GPU in 2016–18) produced outsized returns then multi-quarter mean reversion; plan for similar 20–40% drawdowns on sentiment shocks and size positions accordingly.