Anthropic has publicly filed for a Nasdaq listing under the ticker “ANT,” opening one of the largest prospective technology IPOs to scrutiny with financial disclosures that show rapid enterprise adoption of its AI models alongside enormous computing commitments.
The S-1 registration statement, filed with the U.S. Securities and Exchange Commission on Oct. 26, follows Anthropic’s earlier confidential submission. The company could list as soon as November, subject to the SEC’s review, market conditions and the final terms of the offering. Anthropic was valued at about $965 billion in a private funding round in May 2026, according to the filing materials.
The public document offers a detailed view of a business that has grown far faster than its reported accounting profits suggest. Anthropic generated $16.286 billion in revenue during the first half of 2026, roughly 14 times the amount reported in the comparable period a year earlier. Its results point to commercial AI demand concentrating around paid access to models rather than consumer subscriptions alone.
Api usage supplied four-fifths of first-half revenue
Usage-based API and model calls produced $13.064 billion of Anthropic’s first-half revenue, or about 80% of the total, according to the S-1. Subscription revenue, including paid product plans, accounted for the remaining $3.222 billion.
That mix places Anthropic’s core business closer to enterprise software infrastructure than a conventional subscription application. Companies pay according to their use of Claude models through direct arrangements and cloud platforms, allowing revenue to rise with workloads such as coding assistance, document processing, customer support and internal research tools.
Anthropic reported more than 11,900 customers contributing at least $100,000 in annual revenue as of the end of September. The company counted more than 1,900 customers with annual revenue contributions of at least $1 million, while about 250 customers each generated $10 million or more.
For the first nine months of 2026, Anthropic’s five largest customers generated about 25% of its revenue, with none accounting for more than 10%, the company said. More than half of Fortune 500 companies and over 30% of Global 2000 companies were Anthropic customers that had each contributed at least $100,000 over the prior 12 months.
The figures show strong customer concentration at the enterprise level without dependence on a single buyer. They also indicate that large organizations are moving beyond limited AI trials toward recurring deployments that can produce multi-million-dollar annual spending.
Cloud marketplaces dominate distribution
Anthropic said AWS, Google Cloud and Microsoft Azure together accounted for about 48% of its first-half revenue through their cloud marketplaces. That dependence gives the company access to established enterprise procurement systems and existing cloud budgets, while also tying a substantial part of its sales channel to a small group of technology platforms.
The company’s net revenue retention rate was approximately 900%, excluding personal Claude Pro and Max subscribers. Net revenue retention measures spending by an existing customer base over time; a figure at that level indicates that customers using Anthropic’s enterprise products substantially increased their spending after initial adoption.
Cloud distribution also helps explain the scale of Anthropic’s usage-based revenue. Corporate buyers can access models through infrastructure providers they already use for storage, computing and application deployment, reducing the friction of signing up for a separate AI vendor. For Anthropic, the model extends reach but leaves it exposed to platform fees, commercial negotiations and changes in cloud partners’ own AI strategies.
Accounting loss reflects valuation-linked instruments
Anthropic reported an operating loss of about $693 million for the first six months of 2026. On an adjusted basis, it recorded operating profit of approximately $1.618 billion and an adjusted operating margin of about 10%, after excluding stock-based compensation, non-cash philanthropic spending and certain litigation expenses.
Second-quarter operating profit reached about $974 million, Anthropic said, representing the first profitable quarter on that measure among the periods disclosed in the registration statement.
The company’s reported first-half net loss was far larger, at $80.75 billion. Anthropic attributed approximately $72.4 billion of that figure to fair-value changes in financing instruments, including convertible notes, derivative liabilities and SAFEs, or simple agreements for future equity. These accounting remeasurements are non-cash, but they can grow sharply when a private company’s valuation rises because future claims on its equity become more valuable.
That distinction will be central to how public-market traders assess the IPO. The operating figures present a company approaching profitability after adjustments, while the statutory net loss reflects the cost of financing a rapidly appreciating private business before its shares become publicly traded.
Computing contracts remain a major commitment
Anthropic held approximately $131.7 billion in cash, cash equivalents and other short-term liquid assets at the end of June, according to the filing. The company also disclosed long-term computing agreements containing minimum purchase commitments.
Those commitments are a defining financial risk for large AI developers. Training and operating advanced models requires vast quantities of specialized computing hardware, data-center capacity and electricity. The same infrastructure expense that supports fast revenue growth can create fixed obligations if customer demand slows or if competition compresses model pricing.
The public filing arrives as semiconductor shares rose during the session in which the document became public, with the Philadelphia Semiconductor Index gaining about 2.9%. The move reflected broader attention on the hardware and data-center suppliers positioned to benefit from AI spending, though a single trading day does not establish a direct connection to Anthropic’s planned listing.
For cryptocurrency markets, the filing offers a clearer benchmark for comparing AI-linked narratives with an operating business. Anthropic’s revenue is tied to measurable model usage, enterprise contracts and cloud distribution, while its costs are tied to computing capacity. Tokens associated with decentralized AI projects may attract attention from the same thematic capital, but their valuations remain dependent on their own adoption, token economics and liquidity rather than on Anthropic’s IPO timetable.
See how AI is reshaping trading: explore our AI copy trading guide for data-driven automated strategies.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
