Claude vs ChatGPT: Claude Mobile Revenue Rises 1,000%. Is Anthropic Catching Up To OpenAI?

Claude has not overtaken ChatGPT. But the gap is no longer as comfortable as it once looked. Anthropic is turning Claude from a specialist tool for developers into a serious consumer product, while its enterprise business has grown so quickly that the company now reports a higher revenue run-rate and a higher private valuation than OpenAI. 

OpenAI still owns the larger audience, broader product suite and stronger everyday habit. Anthropic, however, is proving that a smaller audience can be extremely valuable when those users bring Claude into expensive, business-critical work.

Let’s break down what Claude’s 1,000% revenue surge actually means, where ChatGPT still leads, why Anthropic is gaining ground, and how investors should compare the two companies before their potential IPOs. We will also build a simple framework to judge whether OpenAI’s wide moat or Anthropic’s deep moat is more durable.

Key Takeaways

• Claude’s mobile revenue jumped 1,196% to $97.2 million between January and July 2026.
• ChatGPT still leads, earning $312.2 million from mobile users in July.
• Anthropic is catching OpenAI in enterprise AI and coding.
• OpenAI leads in reach, while Anthropic leads in specialised business use.
• Both companies are preparing for potential IPOs.

Claude Revenue Growth Explained: What the 1,000% Surge Really Means

The viral headline needs one important correction. Claude’s total company revenue did not rise 1,000% between January and July. The figure refers only to estimated consumer spending through the Claude mobile app on Apple’s App Store and Google Play.

According to AppMagic estimates reported by Benzinga, Claude generated $97.2 million in mobile consumer revenue in July 2026, up from $7.5 million in January. That is an increase of about 1,196%, or almost 13 times in six months.

ChatGPT generated an estimated $312.2 million from mobile consumers in July. Its revenue increased 28% over the same period, which implies monthly mobile revenue of roughly $243.9 million in January.

These estimates exclude subscriptions purchased on the web, enterprise contracts and API sales. Treat the data as a window into mobile monetisation, not as an income statement for either company.

The clean conclusion is that Claude is catching up rapidly, but ChatGPT still earns more than three times as much from mobile users. Growth tells us who has momentum. Absolute revenue tells us who still has scale.

Claude vs ChatGPT Revenue: The Math Behind Claude’s Growth

The headline becomes more useful when we ask three separate questions. How much share did Claude gain, how much did the actual dollar gap close, and how effectively is each app monetising its audience?

Claude’s share of the two apps’ combined mobile revenue jumped from about 3% to nearly 24%. That is a genuine change in competitive position. Yet the absolute monthly revenue gap narrowed by only about $21 million, or 9%. Claude closed a large percentage of the relative gap because it began from a very small base, but it has not yet seriously erased ChatGPT’s dollar lead.

There is another signal hidden in the July numbers. Claude generated about $5.86 in monthly mobile revenue for every July download, compared with roughly $4.01 for ChatGPT. This is not customer lifetime value because revenue also came from earlier users, but it suggests Claude may attract more power users willing to pay for serious work.

Think of ChatGPT as a crowded supermarket and Claude as a specialist store. ChatGPT serves far more visitors, while a larger share of Claude users may arrive intending to buy something expensive.

That is why the 1,000% story matters. Anthropic is not merely collecting downloads. It is building evidence that its enterprise-first reputation can be converted into consumer spending.

Claude vs ChatGPT Comparison: Revenue, Users, Coding and Enterprise AI

The two products are increasingly similar on the surface. Both can search the web, analyse files, write and run code, remember context, conduct research and work with external tools. The strategic difference lies in where each company built its first stronghold.

OpenAI began with mass consumer adoption and is pushing that audience into work. Anthropic built credibility with developers and enterprises, then began moving outward to consumers.

The dates and definitions are not perfectly matched because these are private companies that disclose selected numbers during fundraising. The table should be read as a current competitive snapshot, not as an audited comparison.

Round One: ChatGPT Still Owns Consumer Distribution

OpenAI reported in March that ChatGPT had more than 900 million weekly active users and over 50 million consumer subscribers. It also said ChatGPT received six times the monthly web visits and mobile sessions of the next-largest AI app. Sensor Tower estimates placed Claude at about 56 million monthly users in the second quarter, versus one billion for ChatGPT.

This lead lowers the cost of launching products. OpenAI can place search, shopping, voice, images, coding, research and agents in front of existing users without needing a separate app for each service.

OpenAI calls this its AI superapp strategy. The company plans to combine ChatGPT, Codex, browsing and agentic capabilities in one product surface. Its March funding announcement argues that familiarity with ChatGPT at home can become a funnel for adoption at work.

This is OpenAI’s widest moat. People often say “ask ChatGPT” when they mean using any AI assistant, much like “Google it.” Changing that habit is harder than winning a benchmark.

Anthropic has made progress. Claude briefly reached the top of major app-store rankings earlier in 2026, and the July revenue data shows that its consumer base is becoming more valuable. But one month of app momentum does not erase ChatGPT’s installed base, global awareness or broader range of consumer tools.

Our view is clear. Claude is now a credible second consumer product, not a fringe alternative. ChatGPT, however, remains the default consumer platform by a very large margin.

Round Two: Anthropic Has the Stronger Enterprise Momentum

The enterprise race is much closer, and some data now puts Anthropic ahead. Ramp’s AI Index, which uses spending across more than 50,000 US businesses, showed Anthropic adoption reaching 34.4% in April 2026, compared with 32.3% for OpenAI, according to Business Insider. Earlier Financial Times reporting showed the two companies almost level in March, with OpenAI at 35% and Anthropic just below one-third.

Anthropic’s own February disclosure was equally striking. More than 500 customers were spending at least $1 million with the company on an annualised basis, up from only 12 two years earlier. The number of customers spending more than $100,000 annually had increased seven times in one year, while eight of the Fortune 10 were using Claude.

OpenAI is hardly weak in business. It reported more than 9 million paying business users in February. By March, enterprise products generated over 40% of OpenAI’s revenue and were expected to reach parity with consumer revenue by the end of 2026.

The difference is where adoption starts. OpenAI often enters a company because employees already use ChatGPT. Anthropic frequently enters through a high-value technical or analytical workflow, then expands into legal, finance, sales, cybersecurity or research.

This distinction matters financially. A general productivity seat can be cancelled if budgets tighten. A model that becomes embedded in a software release process, research pipeline or regulated workflow is harder to remove because replacing it requires new testing, governance and employee training.

Our view is that OpenAI owns the better top-of-funnel, while Anthropic currently has the stronger land-and-expand story. OpenAI gets through the front door because everyone knows it. Claude digs deeper once it gets inside.

Round Three: Claude Code Built Anthropic’s Beachhead

Claude Code is arguably the single most important product in Anthropic’s rise. It became generally available in May 2025 and reached more than $2.5 billion in revenue run-rate by February 2026. Weekly active users had doubled from the start of the year, business subscriptions had quadrupled, and enterprise customers represented more than half of Claude Code revenue, according to Anthropic’s Series G update.

Coding is an attractive starting point for an AI company because the value can be measured. If a coding agent reduces the time needed to fix bugs, migrate a codebase or ship a product, a business can compare the subscription or token cost with engineer hours saved. This is easier to defend than saying an AI chatbot made meetings slightly more productive.

OpenAI is responding quickly. Codex had more than 2 million weekly users by March, five times the level three months earlier, with usage growing more than 70% month on month. OpenAI also bundles coding into the much larger ChatGPT ecosystem, which can help it cross-sell Codex to existing consumers and businesses.

The model comparison is mixed. OpenAI’s published table shows Claude Fable 5 scoring 59.9 on the external Artificial Analysis Intelligence Index, one point above GPT-5.6 Sol at 58.9. Claude led on SWE-Bench Pro, while GPT-5.6 led on some agent, terminal, browser and computer-use tests.

Price strengthens OpenAI’s case. Anthropic lists Claude Fable 5 at $10 per million input tokens and $50 per million output tokens. GPT-5.6 Sol’s standard price is $5 and $30 respectively, before a temporary promotional discount. That makes OpenAI 50% cheaper on input and 40% cheaper on output at the listed rates.

Benchmarks need caution. Vendors choose tests and settings, while real-world performance depends on tools, speed, reliability and human correction. A cheaper model can cost more per completed task if it needs several retries.

Our conclusion is that Anthropic still owns the stronger coding narrative, but OpenAI is attacking it with distribution and price-performance. Claude’s lead is real, not permanent.

Round Four: ChatGPT Has the Broader Product Moat

ChatGPT is no longer just a chatbot. Its product includes voice, image creation, file analysis, memory, projects, scheduled tasks, custom GPTs, deep research, shopping, browser-based work and a growing range of connected apps. OpenAI is trying to become the interface through which consumers and workers access many kinds of software.

Claude has expanded too. Its paid plans include Claude Code, Cowork, Design, Science, Research, Microsoft 365 connections, projects, memory and remote connectors. Anthropic’s strategy, however, is more focused on complex work than on owning every consumer media format.

OpenAI’s moat is breadth. A user can begin with a simple question, generate an image, analyse a spreadsheet, conduct research and ask an agent to act, all under one account. Each added feature gives the user another reason not to leave.

Anthropic’s moat is depth. Claude is optimised for long, difficult workflows involving codebases, documents, tools and organisational context. Its products are designed to make the model useful inside the work itself, not merely before or after it.

Breadth can strengthen consumer retention, while depth can create enterprise switching costs. The winner depends on whether AI value is captured by a universal assistant or models embedded inside specialised workflows.

Round Five: Anthropic’s Multi-Cloud Position Is Underrated

Anthropic says Claude is available on all three of the largest cloud platforms, through Amazon Bedrock, Google Vertex AI and Microsoft Azure Foundry. It also runs on AWS Trainium, Google TPUs and Nvidia GPUs. AWS remains its primary cloud and training partner.

This is valuable because large businesses rarely want a critical AI system tied to only one infrastructure provider. Banks, healthcare companies and governments may already store data in a particular cloud or require a specific security setup. Making Claude available where that data already lives reduces friction.

Anthropic also created the Model Context Protocol, or MCP, an open standard that lets AI systems connect with external tools and data. By December 2025, Anthropic said MCP had more than 10,000 active public servers and had been adopted by ChatGPT, Gemini, Cursor, Microsoft Copilot and Visual Studio Code. Anthropic later donated MCP to a Linux Foundation initiative, making it vendor-neutral.

An open standard can be used by competitors, but creating it still gives Anthropic developer credibility and influence over the ecosystem’s architecture.

OpenAI’s infrastructure advantage is different. It has amassed enormous capital and a wide network across Microsoft, Oracle, AWS, CoreWeave and Google Cloud, with chips from several suppliers. It is betting that more compute will improve models and reduce cost per unit of intelligence.

Anthropic has the cleaner multi-cloud enterprise story. OpenAI has the larger compute and distribution flywheel. Both advantages are expensive, and neither guarantees attractive margins.

OpenAI vs Anthropic: The Financial Race

Comparing the two companies financially is difficult because neither publishes audited quarterly reports. The most reliable figures come from their funding announcements, but even these use revenue run-rates rather than full-year revenue.

A revenue run-rate is the current pace multiplied into a year. Think of it as a speedometer, not an odometer. It tells us how fast the business is moving now, not how much revenue it actually collected over the previous 12 months.

Anthropic’s May funding announcement said its run-rate revenue had crossed $47 billion, up from $14 billion in February. OpenAI’s March announcement said it was generating $2 billion per month, equal to a $24 billion annual pace.

On these snapshots, Anthropic’s run-rate was almost twice OpenAI’s and its valuation was 13% higher. Yet OpenAI’s revenue multiple was much richer, 35.5 times versus 20.5 times.

The gap shows investors are paying for ChatGPT’s audience, brand, ads, commerce, hardware, agents and the possibility that its superapp captures a large part of the AI economy.

Anthropic’s own valuation math is revealing. Between its February and May funding rounds, its reported run-rate rose from $14 billion to $47 billion, an increase of 236%. Its valuation rose from $380 billion to $965 billion, an increase of 154%. Revenue grew faster than valuation, so its run-rate multiple compressed from 27.1 times to 20.5 times despite the company becoming much more valuable.

This does not make Anthropic cheap. A 20 times revenue multiple is still extreme for a capital-intensive, loss-making company. OpenAI’s higher multiple simply requires more future optionality to work.

There is another warning. Anthropic’s May figure and OpenAI’s March figure come from different dates, and private companies may classify partner or reseller revenue differently. Investors should not declare a clean revenue winner until public filings reveal recognised revenue, gross versus net reporting, and the period covered.

Claude vs ChatGPT Moat: OpenAI’s Width vs Anthropic’s Depth

The best way to compare the two companies is not to ask which chatbot gave a better answer this week. Frontier models change too quickly for that. Investors should judge five slower-moving advantages.

OpenAI’s Moat: Width

More users create more feedback, attract developers, justify compute and give OpenAI more surfaces for new products. ChatGPT is the front door.

Its risk is strategic sprawl. Images, search, shopping, ads, hardware, agents and coding all demand capital and management attention. OpenAI is fighting on almost every front at once.

Anthropic’s Moat: Depth

Anthropic found a valuable wedge in coding, built around serious knowledge work and made Claude available across major clouds. Fewer high-spending customers can create enormous revenue as agent usage expands.

Its risk is concentration too. Coding leadership can change with the next model release, and enterprises increasingly use several model providers rather than choosing one permanently. Open models can also handle simpler tasks at much lower cost.

OpenAI has the wider moat, but Anthropic’s may be deeper in workflows it already owns. The IPO filings must show which advantage produces stronger retention and margins.

Anthropic IPO vs OpenAI IPO: Dates, Valuations and Latest Updates

Both companies confidentially submitted draft IPO filings to the US Securities and Exchange Commission in June 2026. A confidential filing starts the review process without immediately revealing the prospectus to the public.

Anthropic appears closer to the starting line. Reuters reported on August 27, citing The Information, that Anthropic plans to reveal its prospectus shortly after the US Labor Day holiday. A listing could follow in late September or early October, although the schedule can still change due to SEC review or market conditions.

OpenAI has not committed to a timeline. When it announced its confidential filing, the company said it may remain private for a while because some operational decisions are easier outside the public market. Filing gives OpenAI the option to move sooner, not an obligation to list immediately.

The Reuters report also said Anthropic may let existing shareholders sell shares in the offering and is considering a lock-up longer than the customary 180 days. These details are not final.

The IPO race matters because the first company to list will set the public-market benchmark for frontier AI. Its revenue multiple, disclosed gross margin and first few quarters of trading could influence how investors value the second company.

What Investors Should Check in Anthropic and OpenAI IPO Filings

The prospectus, not the funding headline, will decide which company has the stronger financial case. Seven disclosures matter most.

1. Recognised revenue versus run-rate. Investors need full-year and quarterly revenue under the same accounting rules.
2. Gross margin after compute costs. High revenue means little if most of it flows directly to cloud and chip providers.
3. Customer concentration. Anthropic’s rapid growth could depend on a limited number of very large API customers. OpenAI may be more diversified, but that assumption needs evidence.
4. Retention and expansion. Net revenue retention will show whether customers spend more over time or switch between models when performance changes.
5. Compute commitments. Take-or-pay data-centre contracts can become a major liability if demand slows or newer chips make older capacity uneconomic.
6. Related-party economics. Microsoft, Amazon, Google, Nvidia and other backers can be investors, suppliers and distribution partners at the same time. The prospectus must reveal who captures the margin.
7. Governance and voting control. Both firms are public benefit corporations with mission-focused control structures. Public shareholders may have less influence than in a conventional technology company.

The single most important number will be gross margin after all cloud, chip and partner payments. Revenue growth proves demand. Gross margin will tell us whether the demand creates durable shareholder value.

The Biggest Risks to Both OpenAI and Anthropic

1. Models May Not Be Durable Moats: Model gaps can close within months. Businesses can route tasks to different providers, while open standards ease switching. The durable moat may be distribution, workflow data, integrations and cost, not one benchmark score.

2. Compute Could Consume the Economics: Both firms need vast data-centre capacity. They must reduce cost per successful task faster than competition pushes prices down, or cloud and chip suppliers may capture much of the profit.

Internal projections reported by The Wall Street Journal in late 2025 suggested Anthropic expected to break even in 2028, while OpenAI expected losses through 2030 because of its broader infrastructure and product ambitions. These were forecasts, not audited guidance, and the businesses have grown dramatically since then.

3. Regulation Can Change Product Access: Both companies operate models with advanced cyber, scientific and agentic capabilities. Governments may restrict access, demand safety tests or impose data-retention rules. Anthropic’s Fable 5 already uses safeguards that redirect some sensitive requests to a less capable model. Stronger safety can build enterprise trust, but it can also limit the product in high-value fields.

4. Copyright Liability Is Still Unresolved: AI companies face lawsuits over the data used to train their models and the content their products generate. Anthropic has already settled a major case involving books and now faces fresh claims from music publishers. OpenAI faces its own copyright litigation. IPO investors will need to examine reserves, indemnities and whether future training requires more expensive licensed data.

5. Governance May Frustrate Public Shareholders: Anthropic’s Long-Term Benefit Trust can ultimately select a majority of its board. OpenAI’s Foundation controls OpenAI Group and can appoint or replace its directors. These structures can protect long-term safety goals, but they also mean management may reject commercially attractive opportunities or capital-allocation choices that ordinary shareholders prefer.

So, Is Claude Beating ChatGPT?

Not overall. Claude is winning the growth narrative, the latest reported revenue run-rate, coding credibility and parts of enterprise adoption. ChatGPT is winning consumer reach, absolute mobile revenue, product breadth, brand awareness and the number of ways it can monetise its audience.

The most accurate verdict depends on the scoreboard.

• If the question is which product more people use, ChatGPT wins comfortably.
• If the question is which company has stronger recent revenue and enterprise momentum, Anthropic currently has the edge.
• If the question is which model is best, there is no permanent answer. Claude and GPT lead different tests, and rankings change quickly.
• If the question is which company has the better investment case, the public prospectuses are essential.

Our strongest view is that Anthropic is no longer merely catching OpenAI. It has already built a different kind of AI leader. OpenAI is trying to become the universal AI interface. Anthropic is trying to become the intelligence layer inside valuable work.

The market may be large enough for both. But their valuations leave little room for both to win only modestly. OpenAI needs its vast reach to produce many new revenue streams and eventually absorb its compute spending. Anthropic needs enterprise usage to remain sticky even when rival models become cheaper or temporarily better.

Claude’s 1,000% mobile revenue surge is therefore not proof that ChatGPT has lost. It is proof that Anthropic can now attack OpenAI on its strongest ground without surrendering its enterprise advantage. The race has changed from one clear leader and one specialist challenger into a contest between a wide moat and a deep one.