Anthropic opened its first day of trading on the Nasdaq at $111.50, 34.3 percent above its $83 initial public offering price, briefly giving the company a fully diluted valuation of about $2 trillion. The stock closed at $97.17, up about 17 percent, and the company ended the day worth roughly $1.77 trillion on a fully diluted basis.
The debut was the largest IPO in U.S. history. Anthropic sold 980 million shares at $83, above the $70 to $78 range it had marketed, Times Now reported, raising $81.34 billion. A concurrent $26 billion private placement from strategic investors, including Google and Nvidia at $10 billion each, brought the total fundraising to about $107.3 billion. The opening valuation edged past SpaceX's June 2026 listing as the largest on record.
The deal
The IPO priced above its range, which is a sign of demand. The deal size and the private placement together made it less a conventional offering than a capital formation event: the company raised money from public investors and from the companies that supply its chips and cloud capacity at the same time.
The private placement included Microsoft, AMD, Broadcom, Micron, and TSMC alongside Google and Nvidia. Those are the companies that build, power, and package the infrastructure Anthropic depends on. Their participation aligns the capital raise with the supply chain, and it also concentrates the shareholder base among strategic partners.
Goldman Sachs and Morgan Stanley led the offering, with a fee reported at about $550 million. The underwriters have an option to sell additional shares, which could lift the proceeds to about $93.5 billion if exercised in full.
What an IPO of this size does to the market
An $81 billion offering absorbs a lot of capital. The deal was priced above its range and still drew enough demand to open 34 percent higher, which tells you how much money was waiting for a piece of the AI trade. The private placement added another $26 billion from strategic investors, so the public offering was only part of the capital raised.
The listing also creates a public benchmark for the sector. Before this week, AI labs were valued in private rounds and secondary transactions. Now there is a daily price, a market capitalization, and a set of financial disclosures that competitors and customers can compare. That changes the conversation about what the companies are worth.
Index inclusion will follow. A company of this size will be added to major indexes, which means passive funds will have to buy the stock. That demand is mechanical and can support the price, and it can also amplify moves in either direction.
The S-1 disclosures
The registration statement gave the public its first detailed look at Anthropic's finances. The revenue growth is the headline: $33.24 billion in the first nine months of 2026, with a third quarter of about $17 billion. The annualized run rate of about $81 billion as of October 31 is the number the company is asking investors to underwrite.
The compute commitments are the other side. Roughly $895 billion to $900 billion in multi-year contracts is a staggering obligation. It is also the cost of doing business in frontier AI, where access to chips and data centers is the constraint on growth. The company is betting that demand will grow into the capacity it has secured.
The GAAP net loss of about $104 billion looks alarming until you read the footnote. Most of it is non-cash charges tied to convertible instruments, not money leaving the bank. The cash burn is the number that matters for liquidity, and the prospectus describes a company with substantial capital and substantial spending.
The valuation
At the IPO price, Anthropic's fully diluted valuation was about $1.5 trillion. The opening trade pushed it to about $2 trillion, and the close left it at about $1.77 trillion. On an outstanding-share basis, the close valued the company at roughly $1.5 trillion, around Tesla's level.
The gap between the opening and the close is the first-day story. A 34 percent pop followed by a 17 percent close means the buyers who got the opening price did well, and the buyers who chased the high did not. That pattern is common in hot IPOs, and it is a reminder that the opening trade is a price set by scarcity, not by a settled valuation.
The comparison to SpaceX is unavoidable. Both companies are capital-intensive, both are led by founders with a public profile, and both drew extraordinary demand. The difference is the business: SpaceX sells launches and connectivity, and Anthropic sells access to AI models. The market is pricing both as category leaders in industries that may define the next decade.
The business
Anthropic reported $33.24 billion in revenue for the first nine months of 2026. Its third-quarter revenue was about $17 billion, and its annualized run rate was about $81 billion as of October 31. Those are extraordinary numbers for a company that was founded in 2021.
The company also carries extraordinary costs. It has roughly $895 billion to $900 billion in compute commitments, the multi-year contracts that secure the chips and data-center capacity it needs. Its GAAP net loss was about $104 billion, most of it non-cash charges related to convertible instruments. A non-cash loss is different from a cash burn, and it is still a number investors have to read carefully.
The business model is capacity-constrained. Anthropic's revenue depends on its ability to serve models to customers, and its ability to serve depends on the compute it has contracted. The commitments are a bet that demand will continue to grow faster than the cost of serving it.
The competition
Anthropic competes with OpenAI, Google DeepMind, Meta, and a set of well-funded challengers. OpenAI is the best-known consumer brand in the category and is reportedly preparing its own listing. Google has its own models and its own cloud. Meta has open-weight models and a distribution channel that reaches billions of users.
The competition is not only about model quality. It is about distribution, enterprise relationships, and the cost of serving. Anthropic's enterprise business has been strong, and its consumer presence is smaller than OpenAI's. The public market will judge whether the enterprise focus is a durable advantage or a narrower niche.
The strategic investors complicate the competition. Google and Nvidia are shareholders and suppliers; Microsoft is a shareholder and a competitor through its OpenAI relationship. A public company has to disclose those relationships, and investors will ask whether the partnerships are arms-length.
The AI trade
Anthropic is the first frontier AI lab to go public. OpenAI is reportedly eyeing a 2027 listing, and the rest of the sector is watching how the public market values a company whose product is a model and whose costs are measured in gigawatts.
The debut is a test of the AI trade. If Anthropic holds its valuation and grows into it, the market has a benchmark for the next AI listing. If the stock fades, the benchmark becomes a cautionary tale. The first-day close is too early to tell, and the first earnings report as a public company will matter more.
The strategic investors add a layer. Google and Nvidia are both suppliers and shareholders, and Microsoft, AMD, and the others have their own interests. A public Anthropic has to manage those relationships in a way a private company did not, because every contract with a shareholder is also a related-party transaction that investors will scrutinize.
What the first earnings will show
The first quarterly report as a public company will be the real test. Investors will look at revenue growth, gross margin, and the pace of compute spending. A company that grows revenue faster than it grows costs is on a path to profitability. A company that spends faster than it earns is buying market share, which can be a good strategy and is not the same as a profitable business.
The report will also show how much of the revenue is concentrated in a few customers. Enterprise AI contracts can be large and lumpy, and a handful of customers can make the difference between a strong quarter and a weak one. The prospectus may disclose some of that concentration; the earnings call will fill in the rest.
The lockup expiration is the other date to watch. When early investors can sell, the supply of shares increases. A stock that holds its price through the lockup has a stronger base of long-term holders. A stock that falls through it has a base of traders.
The fade
The fade from the opening price is not necessarily a bad sign. IPO pops are often driven by limited supply and pent-up demand, and the price settles as more shares trade. A 17 percent first-day gain is still a strong debut by historical standards.
The fade does suggest that the market is not willing to pay any price for AI exposure. Investors are weighing the revenue growth against the compute commitments, the losses, and the competition. The companies that can show a path to profitable growth will command a premium; the ones that cannot will be repriced.
The lockup expiration will be the next test. When insiders and early investors can sell, the supply of shares will rise, and the stock will have to absorb it. The first earnings report will give the market its first look at the company's margins and its spending plans.
A milestone with a question mark
Anthropic's debut is a milestone for the AI industry and for the public markets. It shows that investors will fund a frontier lab at a scale that would have seemed impossible a few years ago, and it shows that the market will reprice quickly when the opening enthusiasm meets the reality of a $900 billion compute bill.
The company has the revenue, the investors, and the attention. The question is whether the business can grow into the valuation. The answer will come from the earnings reports, the compute contracts, and the competition, and it will take more than one day to know.
Primary sources
- Reuters, via Japan News, for the opening price, the IPO size, and the private placement details.
- Livemint, for the $2 trillion opening valuation and the share-price move.
- Investing.com, for the closing price and the fully diluted valuation.
- Times Now, for the offering terms and the underwriter details.