Anthropic filed its S-1 on October 26. It is the first public look at the company's finances. Hover or tap any chart for exact numbers.
30× in five quarters
A year ago Anthropic was doing under $1B a quarter. Now it's doing $11.6B, and the last quarter alone was more than double the one before.
Quarterly revenue
30.7× in five quarters
Show the data
| Quarter | Revenue |
|---|---|
| Q1 2025 | $376M |
| Q2 2025 | $787M |
| Q3 2025 | $1.4B |
| Q4 2025 | $2.0B |
| Q1 2026 | $4.7B |
| Q2 2026 | $11.6B |
$41M of revenue in all of 2023, $386M in 2024, $4.6B in 2025. The first six months of 2026 brought in more than three times all of last year.
First profitable quarter
The usual worry about AI labs is that every dollar of revenue costs more than a dollar of compute. Chips and data centers are expensive, and the more people use the product, the more you spend.
That was true here until this year. Below the dashed line, compute cost more than it brought in.
Revenue per $1 of compute expense
Sub-$1 through 2025; above break-even from Q1 2026.
Show the data
| Quarter | Revenue | Compute | Revenue per $1 |
|---|---|---|---|
| Q1 2025 | $376M | $944M | $0.40 |
| Q2 2025 | $787M | $1.4B | $0.57 |
| Q3 2025 | $1.4B | $2.0B | $0.68 |
| Q4 2025 | $2.0B | $2.9B | $0.69 |
| Q1 2026 | $4.7B | $3.4B | $1.40 |
| Q2 2026 | $11.6B | $5.6B | $2.08 |
Revenue grew far faster than the compute bill.
Indexed growth, Q2 2025 = 1×
Revenue 14.7×, compute 4.0×
Show the data
| Line | Q2 2025 | Q2 2026 | Growth |
|---|---|---|---|
| Revenue | $787M | $11.6B | 14.7× |
| Compute | $1.4B | $5.6B | 4.0× |
Each new dollar of revenue cost less to serve than the last, which is called operating leverage. Q2 2026 ended with $974M of operating profit, the company's first profitable quarter.
The compute line also includes training new models, so the margin on serving customers is higher than this chart shows.
Where the money goes
Over the full first half, costs were still slightly ahead of revenue.
Where the money went, first half of 2026
$16.3B in, $17.0B out. Compute: 53% of costs.
Scroll sideways →
Show the data
| From | To | Amount |
|---|---|---|
| Usage-based | Revenue | $13.1B |
| Subscriptions | Revenue | $3.2B |
| Revenue | Total costs | $16.3B |
| Operating loss | Total costs | $693M |
| Total costs | Compute & infra | $8.9B |
| Total costs | Everything else | $8.0B |
| Everything else | S&M | $3.2B |
| Everything else | Personnel | $3.2B |
| Everything else | G&A & other | $1.7B |
About 80% of revenue is usage-based: businesses call Claude through the API and pay by the token, like a utility bill. Compute eats more than half of all spending. The thin red band is the $693M gap, which Anthropic paid from cash.
The $80.7B paper loss
Most of the $80.7B net loss is an accounting entry.
How the $80.7B loss adds up
90% is an accounting entry. No cash moved.
Show the data
| Step | Amount |
|---|---|
| Operating loss | −$693M |
| Fair-value remeasurement | −$72.4B |
| Other expense | −$8.8B |
| Interest income | $1.1B |
| Income tax | −$21M |
| Net loss | −$80.7B |
Almost all of it is one line: a "fair-value remeasurement." Some early investors put money in through convertible notes and SAFEs, which are IOUs that turn into shares later. When Anthropic's valuation shot up, those IOUs became worth a lot more, and accounting rules make the company book that increase as a loss.
No cash moved. Operating cash flow was about $3.9B positive, and Anthropic held $131.7B of cash and investments in June.
Who pays
The number of customers spending $100K+ a year rose more than 12× in twelve months.
Customers with $100K+ trailing-12-month spend
12.5× YoY
Show the data
| Date | Customers at $100K+ |
|---|---|
| Jun 2025 | 600+ |
| Sep 2025 | 950+ |
| Jun 2026 | 6,000+ |
| Sep 2026 | 11,900+ |
At the top, about 250 customers each spend $10M or more a year.
Customer tiers by trailing-12-month spend
Nested; area to scale.
Show the data
| Tier | Customers |
|---|---|
| $100K+ | 11,900+ |
| $1M+ | 1,900+ |
| $10M+ | 250+ |
Net revenue retention was about 900%. That means the average business customer spent nine times more this year than last.
Revenue is also concentrated.
Who pays, and how (each square = 1% of revenue)
Three cuts of H1 2026 revenue.
80% usage-based
48% via hyperscaler marketplaces
20% from the top two customers
Show the data
| Share of revenue | What |
|---|---|
| 80% | usage-based (mostly API) |
| 48% | via hyperscaler marketplaces (AWS, Google Cloud, Azure) |
| 20% | from the top two customers (unnamed) |
Almost half comes through Amazon, Google and Microsoft's cloud marketplaces, and all three also build competing models. A fifth comes from just two customers. The S-1 says that share is shrinking.
The $637B compute bill
Compute commitments by counterparty
≈$637B aggregate. Blue: H1 2026 revenue.
Show the data
| Company | Amount | Basis |
|---|---|---|
| Broadcom | $161.2B | about, leases |
| SpaceXAI | $120.5B | up to, can cancel |
| $114.5B | at least | |
| Amazon | $110.0B | at least |
| Microsoft | $44.9B | at least |
| Nscale | $44.6B | up to |
| Lambda | $41.3B | about |
Anthropic has signed deals with seven companies for chips, data centers and cloud capacity, worth roughly $637B in total. That's 39 times what it made in the first half of the year. Not all of it is locked in. Some amounts are ceilings and some deals can be canceled. But the part Anthropic must pay no matter what was $265.6B as of June, and most of it runs to 2033 or later.
The bill vs. the bank account
The must-pay part alone is 2.0× the cash. Area to scale.
Show the data
| Item | Billions |
|---|---|
| Compute commitments | $637B |
| Must-pay minimum | $265.6B |
| Cash & investments | $131.7B |
| H1 revenue | $16.3B |
The costs are fixed for years. The revenue is mostly pay-as-you-go, and customers can cut back any month. If demand slows, the contract payments do not.
Bull vs. bear
| The optimistic read | The worried read | |
|---|---|---|
| Margins | The cost to do the same task keeps falling ~10× a year, faster than prices. Margins keep widening. | Open-source models get good enough for routine work and push prices down. |
| Demand | Maybe 1M companies could spend $100K+ a year. Anthropic has ~12K. It has barely started. | Turning a $1M pilot into company-wide spend needs proof of ROI that may not show up. |
| Capital | There is enough cash and borrowing room to fund the contracts for years. | The whole AI industry is borrowing so much it could push up interest rates for everyone. |
| The bottleneck | The real risk is not building enough, as AI agents use far more compute than chat. | The S-1's own risk section is mostly about chips, power and data centers, not customers. |
Risks
The risk factors run more than 90 pages, about a quarter of the filing. Anthropic writes that its models "may behave in unintended, misaligned, or autonomous ways" and mentions "catastrophic or existential" risks that insurers might not cover.
Some are already happening:
- The government fight. The Department of War labeled Anthropic a "supply chain risk" after it refused some military uses. Anthropic lost a 2–1 court ruling in September and asked for a rehearing.
- Copyright. It agreed to a $1.5B settlement over books, with $1.2B still unpaid in June. More suits are pending.
- Control. The founders keep 50.1% of the votes. As a public benefit corporation, Anthropic is allowed to pick the public good over shareholder returns.
What's next
- Q3 revenue. The S-1 leaves it blank. Reuters reports $16.8B, which is still growth but a big slowdown from the quarter before. Part of that may come from Anthropic cracking down on distillation, where others copy its models. Price cuts could speed things back up.
- The IPO price. An updated filing will set the range. Expectations run from an $80B to a $120B raise, which would make it the biggest IPO ever.
- The bill coming due. Most of those compute contracts start in 2026 and 2027. Costs will climb before we know whether revenue keeps pace.
Sources
- Anthropic, PBC Form S-1, SEC EDGAR, October 26, 2026
- Anthropic: S-1 filed with the SEC
- Allocations: Anthropic quarterly revenue, what the S-1 shows
- The VC Corner: Inside the S-1
- FourWeekMBA on the legal matters
- Shattered on compute and safety risks
- Bernama/Xinhua: Q3 revenue report
- All-In Ep. 295: Anthropic S-1 Shocker, Greatest Business Ever? (October 30, 2026)