OpenAI was reported to be generating revenue at a roughly $25 billion annualized pace in late February 2026. That looked like a remarkable lead until Anthropic reported in May that its own run-rate revenue had crossed $47 billion. The newer figure changes the story, but it does not settle who has the larger or healthier business: these are run-rate figures, not audited annual results, and the companies have not published enough comparable financial detail to make a clean ranking.
What the $25 billion figure means
OpenAI’s roughly $25 billion figure was reported as an annualized revenue run rate around late February 2026. In practical terms, that describes a recent pace of revenue scaled to a year; it does not mean OpenAI had already recognized $25 billion over a completed 12-month period. The early-March figure came from secondary reporting, not a complete audited revenue statement from OpenAI. WinBuzzer’s March 6 report also put OpenAI at about $21.4 billion annualized at the end of 2025 and about $6 billion at the end of 2024.
- Annual revenue is revenue recognized during a completed 12-month reporting period.
- Annualized revenue or a run rate takes a current pace and projects it across a year. It can move quickly if usage or sales accelerate or fall.
- Bookings are customer commitments; they may be recognized as revenue over time rather than when signed.
- Consumption revenue, such as usage-based API sales, varies with customer activity and can change materially from one period to another.
- Profit and cash flow account for costs and other financial items. A large revenue run rate says nothing by itself about whether a business is profitable.
Run rate is useful as a directional signal of commercial momentum, but it is sensitive to timing. A recent surge, a large contract, or unusually heavy usage can make a snapshot look stronger than the revenue pace that persists across a full year.
The revenue snapshots changed quickly
| Company and snapshot | Reported amount | What the figure establishes |
|---|---|---|
| OpenAI, end of 2024 | About $6 billion annualized | Secondary-reported run-rate estimate, not audited annual revenue. WinBuzzer |
| OpenAI, end of 2025 | About $21.4 billion annualized | Secondary-reported run-rate estimate, not a completed-year audited total. WinBuzzer |
| OpenAI, late February 2026 | About $25 billion annualized | Secondary-reported run rate; it is not established here as recognized revenue for a completed year. WinBuzzer |
| Anthropic, early March 2026 | About $19 billion annualized | Secondary-reported snapshot, soon superseded by Anthropic’s May announcement. WinBuzzer |
| Anthropic, May 2026 | More than $47 billion run-rate revenue | Anthropic’s company-reported figure; not an audited annual revenue total. Anthropic |
The early-March framing that Anthropic was closing in on OpenAI became stale within weeks. Anthropic said in its May 2026 Series H announcement that its run-rate revenue had crossed $47 billion earlier that month. That is materially above the previously reported $25 billion OpenAI pace, but the two figures cannot be treated as a definitive league table without comparable accounting disclosures.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware match#1 Best Overall
How OpenAI’s business generates revenue
OpenAI’s commercial reach spans consumer ChatGPT subscriptions, business and enterprise offerings, API usage, and coding and developer products such as Codex. It also distributes products and capabilities through strategic and cloud relationships. These channels make the headline number less informative on its own: subscriptions can be recurring, while usage-based sales can fluctuate; enterprise agreements may have different terms and recognition patterns.
In March 2026, OpenAI said enterprise revenue exceeded 40% of total revenue and was on track to reach parity with consumer revenue by the end of the year. The parity statement was a forecast, not an achieved result. The company also reported 3 million weekly Codex users and API activity above 15 billion tokens per minute. Those are company-reported usage and mix figures, not audited segment results. OpenAI’s enterprise announcement
OpenAI’s financing and infrastructure plans are part of the commercial context, not revenue. On February 27, it announced $110 billion in new investment at a $730 billion pre-money valuation. On March 31, it announced $122 billion in committed capital at an $852 billion post-money valuation. These were distinct announcements with different valuation bases and should not be added together or mistaken for operating income. February announcement; March announcement
Why Anthropic’s later figure matters
Anthropic’s reported acceleration was particularly associated with coding and enterprise demand. In February, the company said Claude Code had exceeded a $2.5 billion annualized revenue run rate and that enterprise use represented more than half of Claude Code revenue. It also said business subscriptions for Claude Code had quadrupled since the start of 2026 and that more than a dozen customers had spent over $1 million annually two years earlier. These are Anthropic’s own claims, and the Claude Code figure is for one product rather than the entire company. Anthropic’s Series G announcement
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsThe May claim of more than $47 billion in company-wide run-rate revenue suggests commercial momentum beyond the early-March snapshot. It may reflect coding-agent adoption, enterprise contracts, and access through cloud platforms, but the cited announcement does not provide a detailed revenue breakdown that would establish how much each channel contributed. Anthropic’s $65 billion Series H announcement, at a $965 billion post-money valuation, also demonstrates investor willingness to finance expansion; a financing valuation is not proof of profitability or a guarantee of future returns. Anthropic’s Series H announcement
Why the figures are not a like-for-like comparison
Neither company has supplied a common, audited reporting framework in the announcements cited here. The available information does not establish whether the two run rates use identical revenue-recognition methods, include the same product categories, or treat cloud and reseller economics on the same gross-or-net basis. Differences in timing, usage mix, large customer contracts, and distribution arrangements could all affect the comparison.
Cloud partnerships can make the economics harder to interpret: a model provider may reach customers through a cloud marketplace or rely on a cloud partner for infrastructure, while the channel can affect how revenue and costs appear. A discussion in an SEC-hosted filing highlights broader disclosure questions around AI-lab revenue and cloud relationships, but it is not a definitive accounting comparison of OpenAI and Anthropic. SEC-hosted document
Accordingly, the defensible reading is narrow: OpenAI was reported at about $25 billion annualized in late February; Anthropic later said its run rate exceeded $47 billion in May. Anthropic may have surpassed OpenAI on these reported run-rate measures, but the evidence does not establish a directly comparable ranking of recognized revenue, net economics, or total business value.
Free tools Windows power users keep installed
One-click scans. No signup required.
Revenue growth does not establish profitability
AI companies must pay to develop models and serve them. Training and inference compute, chips, electricity, data-center capacity, cloud contracts, and product development can make costs rise alongside customer use. Coding agents are a particularly important test: frequent, multi-step software tasks may produce valuable recurring workflows, but they can also consume substantial compute. Revenue can grow quickly even if the cost of serving each additional customer or task remains high.
Earlier reporting described both companies as unprofitable and cited a long OpenAI profitability timeline. Those claims are reported information and projections, not independently audited results established by the figures above. The key distinction for investors is that fast revenue growth can improve financing capacity without proving attractive unit economics. To assess durability, investors need information on margins, customer renewals, concentration, cash needs, and the cost of compute—not merely a multiplied monthly pace.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the race means for enterprise buyers
A larger reported run rate is not a product recommendation. For a buyer, the useful comparison is the cost and reliability of completing its own workflows, with the ability to change providers if performance, price, or availability shifts.
- Benchmark the models on representative tasks and compare cost per successfully completed workflow, not just token price or brand recognition.
- Evaluate uptime, privacy, security, administrative controls, auditability, and regional-data requirements against the organization’s obligations.
- Review API and enterprise-contract terms, including rate limits, usage measurement, renewal, cancellation, and price-change provisions.
- Check integration with existing identity, data, monitoring, and cloud systems; include any extra platform or marketplace layer in the cost comparison.
- Keep a practical path to model portability or multi-vendor use, especially for long-term deployments and agentic workloads with unpredictable consumption.
OpenAI’s consumer distribution, broad product range, developer ecosystem, and strategic relationships are meaningful advantages. Anthropic’s reported enterprise and coding momentum shows that consumer scale does not automatically confer enterprise dominance. Both companies also depend on access to capital and computing capacity, so the strength of a vendor’s commercial position should be judged alongside its delivery economics and customer fit.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Funding and an IPO filing are not operating results
Funding rounds can finance compute and expansion while increasing expectations attached to a company’s valuation. OpenAI’s 2026 capital announcements and Anthropic’s Series G and Series H rounds signal substantial investor backing, but capital raised is not customer revenue, and a high valuation does not settle whether the underlying business can earn a return on its infrastructure commitments.
Anthropic said it confidentially submitted a draft S-1 to the SEC on June 1, 2026. The company said any offering would depend on SEC review, market conditions, and other factors. A confidential submission is not a completed IPO, and the announcement did not set an offering size or price. Public filings, if and when available, could provide investors with more detail on financial statements, related-party transactions, channel accounting, and risks. Anthropic’s S-1 announcement
What to watch next
- Comparable financial disclosures: audited statements and clearer treatment of direct sales, cloud channels, and usage revenue would help establish how much the run rates can be compared.
- Revenue quality: renewal and expansion data, customer concentration, and the mix of subscriptions versus variable usage would show whether recent demand persists.
- Serving economics: gross margins and compute costs would reveal whether growth in coding and agent usage improves or pressures profitability.
- Capital and capacity commitments: long-term cloud, chip, power, and data-center obligations affect both the ability to scale and the cost of doing so.
- Customer portability: evidence about switching costs and multi-vendor usage will indicate whether either provider can turn current adoption into durable workflow integration.
OpenAI’s reported $25 billion pace showed how large the market for generative AI products had become. Anthropic’s later $47 billion claim makes the original “closes the gap” headline an incomplete snapshot, not a settled verdict. Until comparable financial statements clarify revenue recognition and margins, the run rates are evidence of momentum—not proof of a winner or a profitable business.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →
