Nvidia Earnings Preview: What Are Traders Watching on 20 May 2026
The world’s most important stock reports on Wednesday. NVIDIA’s Q1 FY2027 print isn’t just another earnings release — it’s a referendum on whether the $700 billion hyperscaler AI buildout is accelerating or approaching an inflection point.
The stock has traded sideways for three months, stuck in a range between $190 and $209, whilst the S&P 500 and Nasdaq have pushed higher. On May 14th 2026, NVDA slowly broke out to be the second largest asset by market cap, sitting at $5.47tn. That either creates a compelling setup or a trap, depending on what Jensen Huang says after the bell.
We have looked at Wallstreet’s earnings consensus, the guidance on capex spending, and the valuation setup. Here is what traders need to know before Nvidia’s earnings.
When is Nvidia’s next earnings report?
Wednesday, 20 May 2026, after market close. The press release typically crosses around 4:20pm ET and the conference call starts at 5pm ET. Fiscally it is Q1 FY2027, the first quarter of the year that ends in late January 2027.
As of May 2026, Nvidia is the largest company by market cap in the entire world, and a surprise in either direction rewrites the AI capital-expenditure thesis for at least one more quarter. Here’s what Wall Street is expecting:
Metric | Q1 FY2027 Consensus | Nvidia Guidance | Implied Beat/Miss |
|---|---|---|---|
Revenue | $78.42B | $78.0B (+/- 2%) | Slight beat baked in |
Non-GAAP EPS | $1.74 | Not guided | Street 12% above whisper |
Data Center Revenue | $73.1B | Not broken out | The number that matters |
Non-GAAP Gross Margin | 74.5% | ~75% (+/- 50bps) | Pricing power intact |
Q2 FY2027 Revenue Guide | $85-87B | To be guided | Must exceed for rally |
Wall Street’s revenue estimate sits $400 million above the midpoint of Nvidia’s own guidance. That is unusual – Nvidia typically sandbags, and the Street usually lands below the guided range. This time, consensus is already in the upper half of guidance, which raises the bar for a “beat.” Goldman Sachs is at $80.05B, a full $2 billion above consensus, which tells you where the optimistic money sits.
The non-GAAP gross margin guide of 75% (+/- 50bps) is the critical quality metric. Nvidia recovered to 75.2% in Q4 FY2026 after the H20 China inventory charge depressed full-year FY2026 to 71.1%. If Blackwell yields and system mix push margins toward the upper end of the range, the stock works. Any compression below 73% and the bears will pounce on “peak pricing power” narratives.
What traders should watch for Nvidia’s upcoming earnings?
1. Data Center Revenue Guidance for Q2 FY2027
Consensus expects Q2 guidance around $85–87 billion. The whisper number is likely closer to $90 billion. A guide above $87 billion sends the stock materially higher. Below $85 billion and the “growth deceleration” narrative takes hold. The bear case centres on the comparison: Q1’s $78 billion guide represented +77% YoY growth, but the comp gets harder from here. Q2 FY2026 revenue was $44 billion. A guide of $87 billion would still be +98% YoY. The issue is sequential momentum, not absolute growth.
2. Gross Margin Trajectory
Nvidia guided Q1 non-GAAP gross margin at ~75% (+/- 50bps). The structural question is whether “mid-70s” holds as Blackwell scales toward full production and Rubin ramps. Colette Kress has been consistent on this message. Any deviation – even 100 basis points – gets amplified. The bears want to see pricing pressure from AMD’s MI350 series and hyperscaler custom silicon. The bulls want confirmation that system-level pricing (not just GPU ASPs) is expanding the margin pool.
3. China and Export Commentary
This is the highest-variance variable. Jensen Huang has excluded China from forward guidance entirely since the H20 restrictions took effect in FY2026. The $4.5 billion inventory charge in Q1 FY2026 remains the reference point. Any updated commentary — whether incremental tightening (further restrictions on remaining eligible products) or a policy reversal (resumption of modified export licences) — will move the stock more than any revenue beat.
Morningstar’s Brian Colello summarised the setup well: “We expect to hear some sort of update regarding sales in China. We anticipate that investors will still seek a beat-and-raise quarter.”
What was Nvidia’s latest earnings?
Nvidia’s Q4 FY2026 results on 25 February were objectively strong. Revenue of $68.1 billion beat consensus by $2.5 billion. EPS of $1.62 beat by $0.08. Data Center revenue hit $62.3 billion, up 75% YoY. The Q1 FY2027 revenue guide of $78 billion was $5 billion above the Street. The stock fell 5.5% the next day.
This is the third consecutive quarter that NVDA has sold off after beating estimates. The pattern is now well established: the market prices in perfection, Nvidia delivers excellence, and the stock drops because the whisper numbers were even higher. The post-earnings moves over the last 12 quarters average +2.3%, but the recent trend is negative — three of the last four prints have been met with selling.
How does valuation actually look into the print?
NVDA Valuation vs Peers
NVDA trades at a forward P/E of approximately 27x on FY2027 consensus EPS of $8.02. That is near the bottom of its three-year range (66x average) and below the Mag Seven median. The PEG ratio sits at 0.66, which implies the market is underpricing the earnings growth trajectory but this only works if the $484 billion FY2027 revenue consensus is realised.
NVDA Analyst Ratings
The analyst consensus is 57 Buy, six Overweight, five Hold, and one Sell. The average price target of $269.95 implies 22% upside from the current $220.78. Wells Fargo’s latest raise to $315 (from $265) anchors the bull case, whilst the Street low of $140 represents the tail-risk scenario.
The short interest sits at 282.9 million shares, or 1.2% of float. That is not elevated by historical standards, which means there isn’t a short squeeze fuel tank waiting to ignite. The institutional picture is more nuanced: Tiger Global trimmed its position by 6% in Q4 2025 but still holds 11 million shares worth ~$2 billion. D1 Capital added 86% to its stake, now at 1.1 million shares. Whalewisdom data shows 52 of 70 major funds hold NVDA, with 18 increasing and 24 reducing positions — a mixed picture that reflects rotation more than conviction.
What are the four hedge fund customers actually doing with their capex?
The four largest customers – Microsoft, Amazon, Alphabet, and Meta – have now all reported Q1 2026 results and disclosed their capital expenditure plans. They plan to spend approximately $720 billion combined in calendar 2026, up 77% year over year.
Microsoft (MSFT): $30.88B in fiscal Q3 capex (up 84% YoY). Full-year CY2026 guidance of $190 billion, well above the $152 billion consensus. CFO Amy Hood attributed $25 billion of the increase to higher component pricing – not expansion, but cost inflation. The company remains capacity constrained through 2026. AI revenue surpassed a $37 billion annual run rate, up 123% YoY. Commercial RPO hit $627 billion, nearly doubled.
Amazon (AMZN): $44.2B in Q1 capex (up 77% YoY), the highest quarterly figure of any company in history. Full-year plan of ~$200 billion. AWS grew 28%, its fastest pace in 15 quarters. The chip business (Graviton, Trainium, Nitro) hit a $20 billion revenue run rate with triple-digit growth. AWS RPO reached $364 billion, up 93% YoY, plus a new $100 billion OpenAI commitment.
Alphabet (GOOGL): $35.67B in Q1 capex, more than doubling YoY. Full-year guidance raised to $180–190 billion, up from $175–185 billion. Google Cloud revenue grew 63% to $20 billion. Cloud backlog nearly doubled QoQ to over $460 billion. Gemini Enterprise paid MAUs grew 40% QoQ. CEO Sundar Pichai confirmed “2027 capex to significantly increase compared to 2026.”
Meta (META): Q1 capex at $19.2 billion. Full-year guidance raised to $125–145 billion, up from $115–135 billion, citing higher memory pricing and data centre costs. Revenue grew 33% to $56.3 billion. The company signed $107 billion in new multi-year infrastructure commitments in a single quarter.
The demand signal is clear: hyperscalers are not pulling back. If anything, they are accelerating. Zuckerberg’s comment on the Meta call crystallises the mindset: “Every sign that we’re seeing in our own work and across the industry gives us confidence in this investment.” Nadella, Pichai, and Jassy have all said variants of the same thing.
What’s the Rubin-Blackwell transition risk?
The risk is whether Rubin extends the total addressable market or starts pulling forward demand that Blackwell would otherwise absorb. Jensen’s own positioning on the Q4 call pointed at the former.
Rubin was unveiled alongside the Q4 release as a platform of “six new chips to deliver up to a 10x reduction in inference token cost vs Blackwell”. The first named customers are AWS, Google Cloud, Microsoft Azure, and Oracle Cloud. Nvidia also introduced BlueField-4, a data processing unit that anchors the “NVIDIA Inference Context Memory Storage Platform”, a new AI-native storage category. And the company flagged a multi-year partnership with Meta covering “large-scale deployment of NVIDIA CPUs, networking and millions of NVIDIA Blackwell and Rubin GPUs”.
Jensen’s framing on the call was direct. “Computing demand is growing exponentially – the agentic AI inflection point has arrived. Grace Blackwell with NVLink is the king of inference today – delivering an order-of-magnitude lower cost per token – and Vera Rubin will extend that leadership even further.” That quote is the single cleanest statement of the two-generation thesis: Blackwell still doing the work today, Rubin widening the moat on inference cost tomorrow.
The risk on the Wednesday call is in the seams between those two generations. If the earliest Rubin orders start arriving in bookings for the second half of FY27, the Street will want to see customer commitments, lead times, and initial allocation ratios. If they arrive fast, the upside is a larger-than-modelled FY28. If they arrive slowly, the near-term concern is Blackwell “digestion”, the phrase the Street has learnt to hate. Listen for it on the call.
How do crypto traders trade AI earnings?
On BitMEX, you can trade Nvidia Equity Perps (NVDAUSDT) with up to 20x leverage. With Equity Perps, traders can use crypto as collateral and trade even when traditional markets are closed. Nvidia reports earnings after the market closes, locking out traditional traders.
Step 1: Deposit Crypto into your BitMEX wallet.
If you haven’t signed up for a BitMEX account yet, we’re currently offering $5,050 worth of trading credits to new users – you can register here.
Step 2: Open NVDAUSDT equity perps from the contract selector under ‘TradFi’.
Step 3: Choose to go long (betting the price will appreciate) or short (betting the price will fall).
Before executing your trade, you can adjust your margin type, leverage, and order type.
Margin Type:
Cross margin (uses your entire available balance as shared collateral)
Isolated margin (caps potential loss to the margin assigned for this trade only).
Learn more about margin types here.
Order Types:
Limit Order lets you name your price and sit in the book.
Market fills instantly at the current best price.
Go long if you expect the pair to rise. Go short if you expect it to fall.
Step 4: Monitor both positions from the position overview tab to understand your key metrics.
Trading Tip: Enable Hedge Mode on BitMEX to hold both long and short positions simultaneously for the same contract. Learn more here.
The bottom line
Nvidia’s Q1 FY27 print on 20 May 2026 is the single most important event of the quarter, with record hyperscaler capex and a high bar for a beat-and-raise. The market has priced a meaningful move, and the three scenarios (bull, base, bear) are clearly defined. For crypto traders, the NVDA Equity Perpetual on BitMEX is the only way to directly access this price action round-the-clock, using crypto as collateral with up to 20x leverage, and bypass the after-hours session that locks out most retail equity traders.
Frequently Asked Questions
When exactly is Nvidia’s next earnings report?
Nvidia’s next earnings report is scheduled for Wednesday, 20 May 2026, after the US market close. The press release will cross at approximately 4:20pm Eastern Time, with the conference call starting at 5pm Eastern Time. Fiscally, this is the Q1 FY2027 print, the first quarter of Nvidia’s fiscal year that ends in late January 2027. The date itself matters because Nvidia’s results now move the entire Magnificent Seven complex and, through it, the S&P 500 and Nasdaq 100. Futures markets reopen in Asia hours after the conference call, so the initial reaction to the print is priced in overnight rather than on the next US cash session.
What is Nvidia guiding for Q1 FY2027?
Nvidia’s own guidance for Q1 FY2027 is revenue of $78.0 billion, plus or minus 2%, implying a range of $76.4 billion to $79.6 billion. That guide is $12 billion above the Street consensus of $66.0 billion, one of the widest guide-above-consensus spreads in mega-cap history. Gross margin is guided at 75.0% non-GAAP, plus or minus 50 basis points. Operating expenses are expected at roughly $7.5 billion non-GAAP, including $1.9 billion of stock-based compensation. The full-year tax rate is guided at 17% to 19%. Crucially, the revenue guide assumes zero Data Center compute revenue from China.
What was Nvidia’s revenue last quarter?
Nvidia reported Q4 FY26 revenue of $68.13 billion, up 20% quarter on quarter and 73% year on year, a company record. Data Center revenue alone was $62.3 billion, up 22% QoQ and 75% YoY, also a record. For the full fiscal year, total revenue reached $215.9 billion, up 65% YoY, and full-year Data Center was $193.7 billion, up 68% YoY. Non-GAAP gross margin in Q4 was 75.2%; GAAP gross margin was 75.0%. Q4 GAAP earnings per share were $1.76; non-GAAP earnings per share were $1.62, beating the Street consensus of $1.54. The company returned $41.1 billion to shareholders during FY26 through buybacks and dividends.
What is Nvidia’s price target from analysts?
Thirty-seven sell-side analysts currently cover Nvidia. The consensus rating is Strong Buy, the highest band on most brokers’ scales. The average 12-month price target is $272.08, implying upside of roughly 21.8% from the $223.39 quote on 13 May 2026. For the full fiscal year, consensus EPS is $6.52, up 34% from the $4.90 posted in FY26. That consensus EPS number is itself supported by 15 estimates and a relatively tight range of $6.00 to $7.29. The one note of caution from the technical side is the relative strength index at 71, which sits inside overbought territory, suggesting the price has run ahead of fundamentals on a short-term basis.
Does Nvidia pay a dividend?
Yes, but it is effectively a token dividend rather than an income stock. Nvidia pays $0.04 per share per year, equivalent to $0.01 per quarter, which works out to a dividend yield of approximately 0.02%. That is not the reason anyone owns the stock. Nvidia’s shareholder-return story is buybacks. In FY26 the company returned $41.1 billion to shareholders, the majority through share repurchases, and still has $58.5 billion of buyback authorisation remaining. For investors optimising for yield, Nvidia is not the right name. For investors optimising for capital appreciation and reinvestment of earnings into the buyback programme, Nvidia’s capital return policy is substantial even if the headline dividend yield is negligible.
Has Nvidia ever missed earnings?
Yes, though rarely. Across the last eight quarters Nvidia has beaten Street consensus on both revenue and EPS in seven out of eight reports. The lone exception was Q1 FY26, reported on 28 May 2025, when EPS came in at $0.81 versus a consensus of $0.87, a miss of $0.06. That same quarter still delivered a revenue beat of roughly $1 billion above consensus. Over the trailing eight quarters, Nvidia’s revenue beat has ranged from roughly $1 billion to $2.5 billion per quarter. The broader pattern, beat-and-raise-with-occasional-hiccup, is precisely what makes the current $12 billion guide-above-consensus so unusual: management has effectively pre-committed to a number well above the Street’s baseline.
How much do Nvidia’s biggest customers spend on AI infrastructure?
In Q1 2026 alone, Nvidia’s four largest public hyperscaler customers spent roughly $110 billion to $115 billion on property, plant and equipment. Microsoft posted $30.88 billion in Q3 FY26 additions to property and equipment, up 84% year on year. Meta reported Q1 2026 capex of $19.84 billion and raised its full-year 2026 guide to $125 billion to $145 billion. Alphabet’s Q1 2026 capex was $35.67 billion, up 107% year on year, and its cloud backlog nearly doubled quarter on quarter to more than $460 billion. Amazon’s trailing-twelve-month property and equipment spend grew $59.3 billion year on year, mostly tied to AI. Annualised, the Big 4 are running at $440 billion to $460 billion of capex.
Can I trade Nvidia earnings on BitMEX with crypto?
Yes. BitMEX lists a Nvidia equity perpetual swap alongside other single-name equities including SPY and QQQ, and the contract is margined with crypto collateral rather than fiat. Leverage on single-name equities is capped at 20x, and the perpetual structure means no quarterly rollover, no forced exit on an expiry date, and access through the Wednesday-night window that cash equity traders in most retail brokerage accounts cannot reach. In practice, this lets you take a view on Nvidia’s print in the hours between the 4:20pm ET release, the 5pm ET conference call, and the next US cash session on Thursday morning, all settled in crypto. The trade-off is the eight-hourly funding rate, which replaces the overnight swap of traditional equity margin.
Disclaimer: This article is for general information and does not constitute financial, investment, or tax advice. Trading derivatives and leveraged products carries risk of loss. Always do your own research and consult with a qualified adviser before trading.
BitMEX Blog

















