Palantir reported second quarter results tonight, and they are exceptional by any measure.
The question we set in advance was singular: is US commercial still accelerating, or is growth leaning back on government contracts? The answer came sharply.
What was reported
| Q2 2026 | Year-over-year | |
|---|---|---|
| Revenue | $1.935 billion | +93% |
| US revenue | $1.573 billion | +115% |
| US commercial | $764 million | +149% |
| US government | $809 million | +90% |
| GAAP income from operations | $912 million (47%) | - |
| Adjusted income from operations | $1.194 billion (62%) | - |
| GAAP net income | $1.062 billion (55%) | - |
| Adjusted free cash flow | $1.220 billion (63%) | - |
| EPS | $0.41 | - |
Revenue growth of 93% is the figure hardest to overstate. A company at an annual run-rate above eight billion dollars that is accelerating rather than decelerating is a rare phenomenon. Sequentially, revenue rose 19% - so this is a quarterly trend, not a one-off jump.
The test we set: US commercial
149% growth, to $764 million, and a further 28% sequentially. Against that, US government grew 90% to $809 million.
The structural fact behind the numbers
US commercial is now almost the same size as the US government business: $764 million against $809 million. For years the central bear argument against Palantir was that it is really a government contractor dressed as a software company. This quarter brings closer the moment that argument stops being true - and at the current pace, commercial overtakes government as soon as next quarter.
And the bookings confirm this is not a single quarter: total contract value closed in the quarter reached $3.373 billion, up 49%. Within it, a record $2.132 billion in US commercial - up 153%. Remaining deal value in US commercial stands at $6.238 billion, up 124%.
The company closed 220 deals of at least $1 million, 98 deals of at least $5 million, and 73 deals of at least $10 million.
The second test: who actually pays for AI software
This is a question we have carried through the whole earnings season, and Palantir is the other side of that coin.
At Reddit, the AI story was roughly 5% of revenue - and the stock collapsed 20.70% on results day. The market did not buy the story because the number behind it was small.
At Palantir the number is $764 million in a quarter, growing 149%. That is no longer a narrative - it is a revenue line. And that is what separates the two articles.
Profitability: more surprising than the growth
Companies growing 93% usually burn cash. Palantir did the opposite.
GAAP net income of $1.062 billion, at a 55% margin. Cash from operations of $1.216 billion, at a 63% margin. On the balance sheet: $9.2 billion in cash and short-term US Treasury securities.
And the Rule of 40 score stands at 155%. That measure adds growth rate to profitability margin, with 40 as the line separating a good software company from an ordinary one. 155 is a number the market almost never sees.
Guidance: raised on every line
- Full-year revenue: raised to $8.150-8.158 billion - 82% growth
- Full-year US commercial: raised to above $3.424 billion - at least 134% growth
- Full-year adjusted income from operations: raised to $4.889-4.897 billion
- Full-year adjusted free cash flow: raised to $4.5-4.7 billion
- For Q3: revenue of $2.160-2.164 billion
CEO Alex Karp framed the release around a single idea: "demand for AI sovereignty has now been unleashed", arguing that customers' competitive advantage "should never become the training data for future models".
What remains open
Valuation. The stock closed at $125.65, up 2.10%, and rose more than 12% after hours to $140.89. Under the rule we work by, a stock's reaction to a report is measured at the close of the following session - not in after-hours trading, which is thin and volatile.
The disagreement in one line
Those who see an exceptional business point to 93% growth that is accelerating, a 55% net margin, free cash flow at a 63% margin, and a US commercial business that has almost overtaken government. All of it is true and all of it is documented.
Those who are cautious point to one thing only, and it is not in the report: the multiple. Palantir trades at one of the highest valuations in the market relative to revenue, one that assumes this pace continues for years. An excellent report does not make a high multiple cheap - it only pushes back the moment it gets tested.
My Angle
A personal opinion of Ilan Abramov - not advice, not a recommendation
I try to separate two questions people tend to blend together.
The first is whether this is a good business. After this quarter, that is not genuinely in dispute. Growth of 93% at an eight-billion-dollar annual scale, with a 55% net margin and real free cash flow, is a combination that barely exists. And what impresses me more than the single number is the pace in US commercial: 149% growth and 153% in contracts closed. Bookings lead revenue, which means the coming quarters are already partly signed.
The second is whether it is a good investment at this price, and that is an entirely different question. Palantir trades at a valuation assuming this story runs for a very long time. A report like this does not resolve that - it defers the test.
What I do take from the quarter: the claim that Palantir is a government contractor in disguise barely stands now. $764 million of commercial revenue against $809 million of government is near parity - and at this pace it flips next quarter. That is a structural turning point, not a good quarter.
