Nvidia has split its stock six times since 2000, most recently a 10-for-1 split with split-adjusted trading beginning in June 2024. One share held before the 2000 split would have become 480 shares. Nvidia could split again if its board decides the share price warrants it, though no rule or threshold requires one.
Nvidia is one of the most frequently split stocks in the semiconductor sector. Each event was a forward split, meaning shareholders ended up holding more shares at a proportionally lower price per share.
Split date | Ratio | One share became | Cumulative factor |
June 2000 | 2-for-1 | 2 shares | 2x |
September 2001 | 2-for-1 | 4 shares | 4x |
April 2006 | 2-for-1 | 8 shares | 8x |
September 2007 | 3-for-2 | 12 shares | 12x |
July 2021 | 4-for-1 | 48 shares | 48x |
June 2024 | 10-for-1 | 480 shares | 480x |
The cumulative factor of 480x describes share count only, not investment return. An investor who held one share through all six events would own 480 shares, but the value of that position depends entirely on how the share price moved over those years. Confusing the two is the most common error in stock split coverage. The official ratios and dates are published in Nvidia's
investor relations records.
The 2024 split was the largest ratio in Nvidia's history and arrived during the peak of investor attention on AI chips. The mechanics were straightforward. Nvidia distributed nine additional shares for every share held, with the distribution completed after the market close on June 7, 2024, and the first split-adjusted trading session on June 10.
Shares had traded near $1,200 before the split. Afterward, the equivalent price sat near $120.
An investor holding 10 shares worth roughly $12,000 held 100 shares worth roughly $12,000 the following session. The share count multiplied by ten; the price per share divided by ten; the position value did not move. Nvidia also restated its historical per-share figures on a split-adjusted basis, which is why older earnings numbers pulled from pre-2024 sources will not match what appears in newer filings.
No. This is the single most important point for anyone researching a split, and it holds regardless of ratio or company.
Market capitalization is share price multiplied by shares outstanding. A split divides one term and multiplies the other by the same number, so the product stays fixed. The same logic applies to valuation multiples, because earnings per share adjusts alongside the price.
Metric | Before a 10-for-1 split | After |
Share price | $1,000 | $100 |
Shares outstanding | 1 billion | 10 billion |
Market capitalization | $1 trillion | $1 trillion |
Earnings per share | $25.00 | $2.50 |
Price-to-earnings ratio | 40x | 40x |
Nothing in the second column represents a discount. A stock that looked expensive on a price-to-earnings basis before a split looks exactly as expensive afterward, which is why split events belong in a separate mental category from the
valuation indicators investors actually use, such as PE, PB, PS and PEG. What a lower nominal price does change is practical access: smaller position sizes become possible without fractional shares, and a standard options contract covering 100 shares carries a smaller dollar value.
The splits followed large increases in Nvidia's share price; they did not create the underlying business value. The more useful causal chain runs through operating performance, especially the expansion of Nvidia's data center business.
The latest reported quarter provides a current example. In fiscal Q2 2027, Nvidia reported revenue of $96.2 billion, up 106% year over year, with Data Center revenue of $89.0 billion, up 117%. Those figures can change quickly, so they belong here as evidence of the operating engine rather than as a permanent growth rate from
Primary source.
The concentration remains important. Data Center represented the large majority of Nvidia's latest quarterly revenue, which makes AI infrastructure demand, customer concentration, and the pace of accelerator deployment central to the business thesis. For the broader company framework,
see MEXC's published Nvidia guideThe causal chain runs in one direction: AI infrastructure spending drives GPU and system demand, which drives data center revenue, which drives earnings, which supports a higher share price, which eventually makes a split administratively convenient. Reversing that order produces bad analysis.
It could, and the honest answer is that nobody outside the board knows. A split requires board approval and, depending on the authorized share count, sometimes a shareholder vote. There is no price level that automatically triggers one.
What can be described is the set of conditions that has preceded splits at Nvidia and its peers:
A high nominal share price. Boards sometimes consider splits after a sustained increase in the per-share price, but there is no universal threshold and modern fractional-share trading has reduced the accessibility argument.
Sustained earnings growth. A rising price supported by rising profits is more durable than one driven by multiple expansion, and boards tend to act after the former.
Options market friction. When a single contract represents a large dollar amount, split discussion tends to resurface among active traders.
Peer behavior. Splits often cluster, as large-cap companies act after comparable firms do.
One factor cuts the other way. Fractional share trading is now widely available at brokers, which weakens the accessibility argument that historically justified splits. A board could reasonably decide the practical benefit no longer outweighs the administrative cost.
Split rumors tend to resurface whenever NVDA sets a new high, and they rarely carry information. The metrics that determine whether a high share price is defensible sit elsewhere.
Data Center revenue growth remains a core signal for AI infrastructure demand. Gross margin shows pricing power, product mix, and transition costs; Nvidia reported a 75.0% GAAP gross margin in fiscal Q2 2027. Free cash flow helps show whether reported profits convert to cash, while guidance provides evidence about demand the company expects to recognize next. These operating signals matter far more than the nominal share price created by a split.
These signals also read across the wider chip complex because Nvidia's demand flows through foundry, memory, networking, and equipment suppliers. For the broader spending framework, see
MEXC's published AI CapEx guide. Current Real U.S. Stock availability on MEXC can be checked at
Stock.
A split changes nothing about the risk profile, and the risks deserve genuine weight.
Valuation compression is the first. When a stock trades on high expectations, results that are merely good can still disappoint. Margin normalization is the second, since the premium margins of a supply-constrained cycle rarely persist once capacity catches up. Customer concentration is the third: a small group of hyperscale buyers accounts for a large share of demand, and several are developing custom silicon to reduce their reliance on external suppliers.
Beyond those sit competition from AMD and other accelerator vendors, export controls limiting sales into China, supply bottlenecks in advanced packaging and high-bandwidth memory, and execution risk around each product generation. Nvidia discloses these in its regulatory filings, which are available through the
SEC's public filing database.
Nvidia's most recent split was a 10-for-1 forward split, with shares distributed after the close on June 7, 2024, and split-adjusted trading starting June 10. It was the largest ratio in the company's history.
Nvidia has completed six stock splits since 2000, in 2000, 2001, 2006, 2007, 2021 and 2024. The cumulative effect turned one pre-2000 share into 480 shares.
No, a split lowers the price per share but leaves market capitalization, valuation multiples and ownership percentage unchanged. Only the share count and nominal price move.
It is possible, but it requires a board decision and no threshold forces one. Wide availability of fractional shares has reduced the practical case for splitting.
A split alone does not change a company's market value, operating performance, or valuation multiples. The economic thesis still depends on the business and the price paid for that business, not on whether the share count is about to change.