
Final fall, Anthropic was taking part in second fiddle to OpenAI. It had a decrease valuation, whereas OpenAI was nonetheless drawing a lot of the eye as the primary mover within the generative AI increase. However the dynamic shifted in late November, when Anthropic released Claude Opus 4.5, which gave the corporate’s Claude Code coding agent a brand new mind and helped elevate it to the standing of “AI killer app.” Arguably, that was the second that set Anthropic on its path towards a forthcoming initial public offering (IPO).
Builders had been utilizing Claude Code to construct software program for a lot of 2025, however the instrument had proven extra promise than actually game-changing outcomes. Opus 4.5 gave Claude Code the intelligence to construct an app or characteristic from finish to finish, primarily based solely on plain-language planning and steerage prompts from the consumer.
Opus 4.5 enabled longer-running brokers and higher planning and execution workflows. Claude Code bought higher at discussing a undertaking with an engineer-user, presenting a plan, incorporating suggestions, after which finishing up a targeted set of multistep duties to finish a software program construct. Anthropic sweetened the deal by imposing fewer utilization caps, which makes a giant distinction for software program engineers who spend their days deploying a number of brokers to construct completely different components of a undertaking.
Simply as necessary, Anthropic modified the way in which software program engineers entry Claude Code. As an alternative of utilizing the instrument via their machine’s command line interface, they might now choose the Code tab within the Claude desktop app, which incorporates its personal built-in terminal. Below that tab, the corporate additionally introduced collectively lots of the assets builders usually use, together with a file editor for viewing and modifying code, code-change assessment home windows, parallel coding classes for various duties, and different instruments. Anthropic turned Claude Code from a terminal/chat instrument into one thing nearer to a full desktop coding surroundings.
“Claude Code is great at fixing all these small bugs from outdated tasks. . . . ” tweeted Pieter Ranges, an influential Dutch entrepreneur and self-taught developer. “[B]efore I’d not have the time to repair these sorts of tasks . . . however now it takes me an hour to do that and it really works once more!”
Claude’s beachhead
The brand new unified interface within the Claude desktop app invited software program engineers to do their nonsoftware improvement work in the identical place. With easy accessibility to Claude chat and CoWork, engineers may use Claude for extra normal duties, together with conducting analysis, accessing firm information, composing emails, and styling displays.
The interface additionally accommodates staff who code solely sometimes, maybe to slap collectively a prototype app to current to colleagues, however who profit from doing a lot of their day-to-day work with assist from Claude chat and CoWork’s information connections, abilities, and workflows.
Acceleration
Regardless of the motivation, enterprises are certainly selecting Anthropic, and the corporate’s accelerating enterprise momentum will be traced again to Claude Opus 4.5. On the finish of 2025, Anthropic reported simply $9 billion in annualized income run price. (ARR extrapolates a current month’s income out to a full yr.) That quantity grew quickly in 2026 as Claude Code proliferated. The corporate reported $14 billion in ARR in February and $30 billion in April. A Could Reuters report pegged the corporate’s ARR at $47 billion. For context, OpenAI said on the finish of March that it was producing about $2 billion per thirty days in income, which works out to ARR of $24 billion.
Anthropic reportedly said that greater than 1,000 firms at the moment are paying over $1 million per yr for Claude, and that this quantity has greater than doubled since February 2026. The corporate has introduced main deployments with PwC, Allianz, Snowflake, Accenture, Deloitte, and IBM. Anthropic expects to make an working revenue of $559 million through the quarter ending in June, on income of $10.9 billion, according to The Wall Street Journal.
Whereas Anthropic doesn’t launch uncooked consumer numbers for Claude Code, survey information suggests the instrument has loads of room to develop. An early-April JetBrains survey of 10,000 software program builders globally discovered that 29% used GitHub Copilot, whereas solely 18% used Claude Code, tied with Cursor.
Stage is about
On Monday, Anthropic submitted a confidential draft registration assertion, or S-1, to the Securities and Alternate Fee (SEC). Below U.S. guidelines, the SEC can assessment IPO-seeking firms privately, permitting an organization to delay making detailed financials public till later, nearer to the precise itemizing.
Anthropic’s may find yourself being the second of three large artificial-intelligence-related IPOs in 2026. SpaceX has already filed, and OpenAI is anticipated to file later within the yr. Each Anthropic and OpenAI have an opportunity to IPO at trillion-dollar-plus valuations, placing them among the many largest tech IPOs in historical past. Anthropic simply final week raised $65 billion in new financing at a valuation of $965 billion, together with the brand new cash.
Specialists should not frightened about whether or not enterprise software program engineering teams will undertake Claude Code. They imagine the demand is actual and rising. The extra severe menace, for each Anthropic and its clients, is the price of utilizing the instrument. A part of Claude Code’s enchantment is that it develops a deep understanding of code bases and complicated coding issues. That requires the agent to cause throughout massive context home windows, which implies utilizing a number of tokens, the chunks of textual content, information, and code it processes.
Company customers, together with Uber, are already running up big bills as engineers max out on tokens. Anthropic, in the meantime, is looking for further computing assets to course of all of the tokens being generated by Claude Code customers. The corporate is even paying Elon Musk $1.25 billion a month for the usage of the xAI/SpaceX Colossus 1 information heart in Memphis.
When Anthropic’s prospectus reveals up, possible someday this summer time, we’ll lastly be taught extra about the actual profitability of promoting Claude Code.