SemiAnalysis Latest Report: Anthropic's Third-Quarter Revenue May Top $1 Billion

TradingKey - A recent in-depth report released by research firm SemiAnalysis reveals that AI large model company Anthropic is expected to achieve GAAP EBIT of over $1 billion by the third quarter of 2026, with its annual recurring revenue (ARR) surging from $9 billion at the end of 2025 to over $60 billion. SemiAnalysis believes that if Anthropic continues to execute its strategy efficiently, it has the potential to become the first company to reach a market capitalization of $6 trillion.

[Image source: SemiAnalysis official website; Data source: The Wall Street Journal and SemiAnalysis tokenomics model]

Claude Code drives enterprise adoption, while the API model raises the revenue ceiling.

The most direct engine behind this explosive performance growth is Claude Code, the AI programming tool launched by Anthropic. The report shows that Claude Code currently accounts for over 7% of all code commits on GitHub, and the company's ARR has achieved exponential growth within a few months.

Unlike the path that relies on C-end subscriptions, about 75% to 85% of Anthropic's ARR comes from its usage-based API business, while consumer subscriptions account for only 5% of total ARR. During the same period, its main competitor OpenAI still derived over 65% of its revenue from the subscription model.

The difference between these two paths lies in this: the API model has no revenue ceiling per user. The more agentic workflows enterprise customers deploy, the more token consumption and revenue will grow alongside them, allowing for expansion without the need to constantly acquire new users. Data disclosed by Anthropic's CFO shows that the company's Net Revenue Retention (NRR) reached 500%, and the customer cohort contributing $30 billion in ARR this year contributed only $2 billion a year ago.

Gross margin recovers to the 60% range; inference efficiency holds the key

The divergence in business models is directly reflected in financial data. SemiAnalysis estimates that Anthropic's overall gross margin has rebounded from negative 94% in 2024 to the mid-60% range, with the gross margin of its API business exceeding 80%.

The core driver behind the substantial improvement in gross margin is the enhancement of inference efficiency. Measured by ARR per megawatt of computing power, this metric for Anthropic is projected to reach $60 million later this year, up from just $16 million nine months ago. Since inference computing costs are largely fixed, once the volume of tokens processed per unit of computing power or the pricing increases, the marginal profit margin will approach 100%.

The report also draws a comparison between Anthropic and OpenAI: if both companies reach $100 billion in ARR, OpenAI's cost of services would be higher due to supporting its massive free user base, resulting in approximately $25 billion less in gross profit than Anthropic. This gap directly impacts how much both parties can invest in training next-generation models. Anthropic's EBIT margin reached 36% in the second quarter of 2026, and the report forecasts that by 2028, Anthropic's cumulative EBIT will exceed OpenAI's by $250 billion.

As Anthropic's IPO window approaches, the industry's focus shifts from "scrambling for computing power" to "calculating returns."

This financial data was released shortly after Anthropic confidentially filed for an IPO on June 1. SemiAnalysis believes there is a certain urgency to going public at this time: tech giants have already completed massive equity financing rounds, the window for the capital markets is narrowing, and Anthropic needs to make an early public debut to seize the initiative in the capital race.

At the same time, the rise of Anthropic is altering the overall logic of the industry. The brute-force race of "grabbing computing power at all costs" over the past two years has reached an inflection point. Meta was recently rumored to plan the launch of a cloud infrastructure business to rent out AI computing power; other media reports indicate that Anthropic has initiated early preparations for in-house AI chip development and has entered discussions with Samsung Electronics regarding potential manufacturing collaboration. These movements all demonstrate that the industry's focus is shifting from the scale of capital expenditure to the return on capital.

What this SemiAnalysis report is actually saying comes down to one thing: the next phase of AI commercialization will belong to companies that can turn model capabilities into sustainable, high-gross-margin cash flows. Anthropic has proven this business model is viable with quarterly profits in the range of $1 billion, which not only puts it ahead of the pack but is also redefining how the entire track should be valued.