CoreWeave's Revenue Jumped 168%. Wall Street's Question Is About Funding Costs

AI demand is proven. What matters now is how much of the GPU rental income is left for shareholders after financing costs.

The 20-second read

1
Results were stronger than expected. CoreWeave's third-quarter 2026 revenue was $3.661 billion, up 168% from a year earlier. Adjusted operating margin also rose to 10% from 5% in the prior quarter.
2
But the financial burden remains. Net loss for the same quarter was $709 million. Full-year capital spending is forecast at $37 billion to $39 billion. There is a sizable gap between growing AI demand and returns for shareholders.
3
The most interesting change is the value of older GPUs. Rental prices for existing Ampere and Hopper GPUs rose 30% to 50% from the prior quarter. This suggests the company may be able to earn extra revenue without buying new equipment.

The number that matters more than revenue: a 10% operating margin

CoreWeave's third-quarter results showed that demand in the AI computing market remains strong.

Revenue was $3.661 billion, about 2.7 times the level a year earlier. Adjusted operating income was $379 million, well above the top of the company's own guidance of $260 million.

The adjusted operating margin improved from about 1% in the first quarter to 5% in the second and 10% in the third. Operating leverage, where efficiency improves as revenue grows, is starting to show.

There is a caveat. The adjusted operating margin a year earlier was 16%. This quarter's improvement is more accurately read as a recovery from the prior quarter.

Revenue growth, year over year: +168%. Driven by expanding demand for AI computing.

GAAP net loss: $709 million. Reflects financing costs and depreciation.

The company raised its full-year revenue outlook to $13 billion to $13.3 billion. It also adjusted its capital spending outlook to $37 billion to $39 billion.

The fact that earning more still requires putting in more capital has not changed.

Why does Wall Street read the same results differently?

JPMorgan praised CoreWeave's execution and pricing power and kept its overweight rating and $125 price target.

Morgan Stanley is more cautious. It wants to see the funding plan turn into completed deals. Bernstein acknowledged the operating improvement but questioned the long-term economics of the business model.

Notably, all of them focus less on AI computing demand itself than on how favorable the financing terms are on which CoreWeave can turn that demand into a business.

The company's new funding strategy targets that very issue.

CoreWeave plans to avoid issuing high-interest unsecured bonds. It intends to lean on loans secured by customer contracts, customer prepayments and equipment-vendor financing.

The company expects about $9 billion in customer prepayments and deposits ahead. It is also pursuing new loans at roughly 7% to 7.5% to cover more than 90% of the related capital spending.

If the deals close on the expected terms, the funding burden could ease considerably. For now it is a plan. Actual financing costs can be judged only after the loans are drawn.

The most interesting signal came from older GPUs

The part of this report most worth watching is not the newest GPUs but the profitability of those already installed.

According to management, third-quarter rental prices for Ampere and Hopper-generation GPUs rose 30% to 50% from the previous quarter. That suggests older GPUs can still be re-contracted at fairly high prices.

CoreWeave expects about 130,000 GPUs to come off existing contracts in 2027 and about 170,000 more in 2028. At current rental prices, the company says, those units could create a meaningful amount of additional annualized revenue.

A significant business change sits underneath this.

Buying new GPUs takes huge upfront capital. If existing equipment keeps being rented after it has paid back its cost, it can generate cash flow without new purchases.

That is a key variable in judging CoreWeave's business model.

Today's high rental prices will not last forever. As technology advances and supply expands, the economic value of older GPUs could fall again.

What counts is not a GPU's technical life but its economic life, the period in which it can actually earn money.

A 62% EBITDA margin, so why a loss?

CoreWeave's adjusted EBITDA margin in the third quarter was 62%, a high figure.

Over the same period, depreciation was $1.889 billion and net interest expense was $846 million. The result was a net loss of $709 million.

Adjusted EBITDA excludes items such as equipment depreciation and financing costs. For a company that needs heavy capital spending and borrowing, that number alone makes it hard to judge the economic gain to shareholders.

The company expects fourth-quarter interest expense of $1.05 billion to $1.15 billion.

Operating profit is improving, but so is interest expense. In coming reports, the key question is which of the two grows faster.

How were CoreWeave's third-quarter results?
  • Revenue was $3.661 billion, up 168% from a year earlier.
  • Adjusted operating income was $379 million, above the $260 million top end of the company's own guidance.
  • However, the net loss on a GAAP basis was $709 million.
CoreWeave has a 62% EBITDA margin. Why is it still losing money?
  • Adjusted EBITDA is a measure that excludes items such as depreciation and financing costs.
  • In the third quarter, depreciation was $1.889 billion and net interest expense was $846 million.
  • These costs outweighed profit and led to a net loss.
Why does Wall Street view CoreWeave differently?
  • Analysts broadly agree that demand is strong. Where they differ is on how well that demand can be turned into business on favorable financing terms.
  • JP Morgan kept its overweight rating, citing execution and pricing power.
  • Morgan Stanley wants to see whether the financing actually closes as real deals. Bernstein raised doubts about the economics of the long-term business model.
Why is the rise in older GPU rental prices an important signal?
  • Management said third-quarter rental prices for Ampere and Hopper GPUs rose 30% to 50% from the prior quarter.
  • If equipment whose cost has already been recovered keeps getting rented, it can generate cash flow without buying new equipment.
  • But it is not yet clear whether the high prices will last. They could fall again as technology advances and supply expands.
What is CoreWeave's new financing plan?
  • The company plans to avoid high-interest unsecured bonds and use loans backed by customer contracts, customer prepayments, and equipment supplier financing.
  • The new loan would cover more than 90% of the related capital spending at an interest rate of about 7% to 7.5%.
  • It is still at the planning stage. The actual financing cost can only be confirmed after the loan is made.

What to watch

Fourth-quarter interest expenseThe company expects $1.05 billion to $1.15 billion. Watch whether the actual number falls within this range.
10% adjusted operating marginWatch whether it rises again next quarter. Compare it with how fast interest expense is growing.
Loan termsWatch whether the deal to cover more than 90% of capital spending at about 7% to 7.5% actually closes.
Older GPU rental pricesWatch whether the 30% to 50% rise for Ampere and Hopper continues. Pricing when about 130,000 units come up for renewal in 2027 is also a variable.
CoreWeave is being valued less on AI demand than on how cheaply it can finance that demand.

Insight Times Editorial Desk

More by topic

Starting with this story's topics, then what readers are following now

Investor psychology, decision-making and books on investing

GPU

This story's topic73 storiesSee all →

A chip built to draw graphics, whose ability to run thousands of identical cal…

Why Nebius Is Keeping Some 2027 AI Capacity Off the Shelf

Nebius mixes long-term contracts, short-term premiums and auctions. Holding back capacity pays only if the servers stay busy, and it is not yet clear the pricing will last.

Tech

Who Hits $1 Trillion Next: Reading AMD and ASML by Different Rules

One company is chasing Nvidia's GPU empire. The other sells the machine that draws the chips for every advanced chipmaker, including Nvidia. Both get floated as 2028 trillion-dollar candidates, but the numbers investors need to watch are entirely different.

Investing

Is Space Cooler for AI? Why Musk Wants Data Centers in Orbit

The next limit on AI may be power and cooling, not GPUs. Elon Musk is looking to Earth orbit for an answer, but the test is cost per hour of compute.

Tech

Musk Sees the Next Phase of AI Infrastructure: Who Can Switch GPUs On, Fast

If neoclouds struggle under financial pressure, AI demand may not shrink. It may simply favor the few operators with the capital and execution to serve it, a shift that could matter for SpaceX and Tesla.

Tech

Marvell's $115 Billion Lumentum Deal: Is AI's Next Bottleneck Light, Not Chips?

Marvell's purchase of Lumentum points to a contest beyond GPU speed, in data transfer and power efficiency. Whether the price is right is not yet confirmed.

Tech

Nvidia and Palantir Put AI Inside the Supply Chain - The Real Shift Is That It's Learning Judgment

A single Vera Rubin rack holds 1.3 million parts. The real breakthrough in Nvidia and Palantir's new supply chain AI isn't automating that complexity, it's capturing the judgment calls that used to live only in planners' heads.

Tech

The Next AI Chip Race Is About How Chips Connect

Applied Materials will lift its stake in Dutch equipment maker BESI from about 9% to 20%. The bet is less about control than about a new way to link chips.

Tech

Earnings Report

This story's topic43 storiesSee all →
A person at a kitchen table looks at a food delivery app on a phone beside a laptop and a DoorDash delivery bag

AI Was Supposed to Do Our Shopping. DoorDash Got Only Several Hundred Orders From It

DoorDash logged 993 million orders in the third quarter, yet outside AI agents placed only several hundred in total. Orders built through its own in-app AI ran about 50% larger.

Tech

Amazon's AI Business Is Booming. Why Did the Stock Slide?

AWS revenue rose 39% and operating margin hit a record 13.9%. But capital spending now exceeds the cash Amazon generates, and that gap is what investors are pricing.

Tech

Samsung Posts 107 Trillion Won Profit, TSMC Sales Jump 50%. Why Did the Shares Fall?

The AI chip supercycle has not ended. But the market's question has changed: not how fast earnings grow, but how long today's enormous margins can last.

Tech

Two Companies Drive 34% of S&P 500 Earnings Growth as AI Pays Sellers First

Nvidia and Micron generate a third of expected earnings growth in US large caps. Apollo says the financial effect of AI is not yet visible in the S&P 493.

Investing

Tesla Energy's 13.7GWh Paradox: Expectations Ran Faster Than Deployments

Q2 deployments recovered but margins collapsed. Q3 set a record for deployments and still missed estimates. The issue is not the size of AI power demand but how fast it converts into Tesla revenue and profit.

Tech

Apple and Palantir Beat Estimates, but Two Questions Remain

Apple has to show how much of its revenue turns into profit. Palantir has to show how long its explosive growth can last.

Investing

SK hynix's 80% Margin Matters Less Than Its Pre-Sold 2028 Output

SK hynix earned 83.3 trillion won in operating profit in one quarter. The number long-term investors may care about more is that over half of 2027 and 2028 capacity is already under long-term contracts.

Tech

A building that houses servers and networking gear and supplies them with powe…

AI ran out of memory before it ran out of GPUs

The HBM bottleneck is spreading into DRAM prices, server costs and the price of the next laptop or phone.

Tech

White House Swaps AI for "SI." The Real Investment Signal Is Infrastructure, Not the Name

An executive order renames artificial intelligence "Super Intelligence" and six frontier labs sign a voluntary safety accord. What investors should watch is the data center, power, nuclear and chip policy now gathered under one national-competitiveness frame.

Tech

Weekly Check: The AI Bull Market Isn't Over, It's Just Gotten Harder

A 5% 10-year yield, 470 gigawatts stacked up in Texas, Chinese DRAM, and whether data centers actually get switched on are starting to decide who wins the AI trade through 2027.

Tech

Is Memory Peaking? In the AI Era, Watch the Bottleneck, Not the Demand

Demand still outruns supply. But memory stocks can peak before demand does. The real question is how long HBM4, server DRAM, CXL and Big Tech capex keep the cash flowing.

Tech

SoftBank Bet $64.6 Billion on AI. Why Is the Market Pricing Default Risk?

Believing in AI's future is not the same as backing a company that can survive until it arrives. SoftBank's credit spike raises that question for long-term investors.

Tech

If You Only Watch Tesla, You May Be Missing Half the AI Story

Nvidia builds the AI brain. Who supplies the electricity to run it? Six power companies Tesla investors should know, and why Tesla's storage business is just one link in the chain.

Investing

Why AI Is Moving Back to the Laptop: Anduril and Qualcomm's BlueChip

Anduril and Qualcomm's new BlueChip laptops point to a bigger question: where should AI run? Production and sales are still unproven.

Tech

NVDA

101 storiesSee all →

Box Shows AI Its Manuals, and the AI Starts Redesigning the Workflow

Box Architect, built with NVIDIA, moves AI from reading documents to designing how work gets done. It has not yet reached beta, so the real-world payoff remains to be seen.

Tech

Tech Selloff: Is the Bull Market Over? Wall Street's Four-Year Chart Says Not Yet

The uptrend has survived five double-digit pullbacks in four years. This time, AI companies' debt and cash flow are the new variables.

Investing

Nvidia Moves From GPUs to CPUs as AI Agents Reshape a $200 Billion Chip Market

A customer's early test found Nvidia's Vera CPU handled about twice the AI agent workload of its existing chips. The bigger story is where the money in semiconductors is heading as AI shifts from answering questions to doing work.

Tech

Nvidia's $500 Billion Experiment: Wall Street Asks Who Backs the GPU, Not Just the Chip

The real issue in Nvidia's financing platform is not chip sales. It is how far Wall Street will trust GPU resale prices and rental income three to four years out.

Tech

China's AI Labs Went Open Source Because They Had to, Not Because They Wanted To

A GPU shortfall pushed Chinese AI labs toward efficiency and open weights, and that constraint is now reshaping the global competition over who controls AI's basic infrastructure.

Tech

What Jensen Huang sold at the G20 was not GPUs. It was national infrastructure

Nvidia's message is that AI should be laid down like electricity, roads and the internet. After the hyperscalers, the buyers are governments, telcos, regional clouds and industrial data centers.

Tech

The Cheaper Tokens Get, the Bigger Your AI Bill Gets

As AI shifts from training to inference, the token is becoming an industrial commodity. Falling prices are not cutting demand, because agents and automation consume tokens at a scale no human chat session ever did.

Investing