Palantir Q2 2026: growth accelerated to 93% on a ~$7.7B run rate

Most companies decelerate as they get bigger. That is not an opinion, it is arithmetic: the larger the base, the more absolute revenue each percentage point requires.

Palantir went the other way.

	Q2 2026	Consensus
Revenue	$1.935B	~$1.80–1.81B	beat
Adjusted EPS	$0.41	~$0.35	beat
US government revenue	$809M	~$731M	beat
US commercial revenue	$764M	~$716M	beat
GAAP operating income	$912M	—	47% margin
Adjusted operating income	$1.194B	—	62% margin
GAAP net income	$1.062B	—	55% margin
Adjusted free cash flow	$1.220B	—	63% margin

Revenue grew 93% year over year and 19% sequentially. Last quarter it grew 85%. A company annualising near $7.7 billion posted faster growth than it did three months earlier, and it did so while expanding adjusted operating margin from 46% to 62%.

Full-year revenue guidance went from $7.65–7.66 billion to $8.150–8.158 billion — a raise of roughly half a billion dollars in one quarter, on a business the company was already guiding to grow 71%. Third-quarter revenue is guided to $2.160–2.164 billion.

Did Palantir beat its own guidance in Q2 2026?

Yes. Revenue of $1.935 billion came in above the top of the $1.797–1.801 billion range Palantir guided to in its May 4, 2026 8-K, and adjusted income from operations of $1.194 billion was above the top of the $1.063–1.067 billion it guided to in the same release. Against the year-ago quarter, revenue grew 93% from $1.004 billion, and GAAP diluted EPS was $0.41 against $0.13. The full-year revenue outlook moved from $7.650–7.662 billion in May to $8.150–8.158 billion.

The only number that really matters here

Not the beat. Palantir has beaten consensus for eight consecutive quarters, so a beat on its own tells you little.

The number that matters is 93% against 85% — the direction of the second derivative. At Palantir’s scale, growth reaccelerating is close to unheard of. It means the addressable demand is not being consumed faster than it is being created, which is the central question for any hypergrowth business.

Everything else in the release is downstream of that:

• Rule of 40 at 155% (93% growth + 62% adjusted operating margin), against 145% last quarter and 94% a year ago. The convention is that 40 is good. This is not a variation on the theme; it is a different category of financial profile.
• Adjusted free cash flow of $1.220 billion on $1.935 billion of revenue — a 63% margin, with cash from operations of $1.216 billion. The growth is self-funding, behind $9.2 billion of cash and short-term Treasuries and a balance sheet carrying no borrowings at all.
• Net dollar retention of 157%. Existing customers spent half again as much as they did a year ago.

The mix quietly flipped

This is the structural change in the quarter, and it got less attention than the headline.

	Q2 2026	YoY
Total commercial	$945M	+110%
Total government	$990M	+79%
US commercial	$764M	+149%
US government	$809M	+90%

Commercial revenue is now essentially level with government revenue, and growing faster on both a total and a US basis. US revenue overall reached $1.57 billion, up 115%, and is 81% of the company.

A year ago the fair description was a government contractor with a promising commercial business. It is now, in round numbers, half a commercial software company. That matters because it changes which risks apply. Appropriations cycles, procurement delays and administration priorities matter less to the marginal dollar than they used to. Competitive displacement and the durability of enterprise AI budgets matter more.

Investors who bought Palantir as a defence-adjacent position and investors who bought it as an AI software position are now holding measurably different companies than each believed. Neither has been told; the mix simply moved underneath them.

The growth is contracted, not forecast

The distinction worth drawing on any hypergrowth report is between revenue a company hopes for and revenue it has already signed.

• Total contract value booked in the quarter: $3.373 billion, up 49% — against $1.935 billion recognised. Bookings ran well ahead of revenue.
• US commercial TCV: $2.132 billion, up 153%, a quarterly record.
• US commercial remaining deal value: $6.238 billion, up 124% year over year and 27% sequentially.
• 220 deals closed above $1 million, of which 98 above $5 million and 73 above $10 million.
• Customers: 1,049, up 24%; commercial customers 870, up 26%; US commercial customers 653, up 35%.

Remaining deal value growing 27% in a single quarter is the part that supports the guidance raise. It is not proof that growth persists — deal value can be re-phased, and it converts on the customer’s timetable, not the company’s — but it means the next few quarters are substantially underwritten by contracts that already exist.

How long can the growth last?

Because that is what the figures cannot answer on their own. Growth of 93% on a base near $7.7 billion annualised is the result; the open question is the duration.

The result is growth, not its duration: what a holder is underwriting is how long the rate persists, not whether it exists.

That is precisely the horizon management addressed. CEO Alex Karp called the quarter “otherworldly” and told CNBC the growth “looks like this is going to go on for at least another 18 months.” Read carefully, that is a specific and bounded claim.

One quarter, even one this good, moves the estimate of multi-year growth only slightly. The record so far is acceleration, and the guide is built on it.

What is genuinely strong, stated plainly

The operating record, stated plainly:

• Acceleration at scale. 85% → 93% at a ~$7.7 billion annualised run rate. Very few companies in history have done this.
• 62% adjusted operating margin at 93% growth. Companies growing this fast normally consume cash. This one converts 63% of revenue to adjusted free cash flow.
• Segment beats on both sides. US government beat consensus by roughly $78 million and US commercial by roughly $48 million. The upside was not one lumpy contract.
• 157% net dollar retention with 35% growth in US commercial customers — expansion and new logos, not one substituting for the other.
• A guidance raise of roughly $500 million, plus adjusted free cash flow guidance of $4.5–4.7 billion. Companies do not raise like this into an expected slowdown.

The bear case, framed honestly

It is not that the numbers are wrong or the demand is fake. It is narrower:

1. Concentration. US revenue is 81% of the total, and US government $809 million of it. Two related exposures — one country, one buyer type — dominate.
2. Comparison base. 149% US commercial growth laps $306 million. As the base compounds, the same absolute wins produce smaller percentages, and percentages shrink even when the absolute wins do not.
3. The 18-month framing. Management’s own stated visibility is bounded, and growth rates beyond it are not guided.
4. Margin at 62% is now the comparison. Extraordinary margins become the baseline that next year is measured against. Hiring to serve $6.24 billion of remaining deal value is not free.

None of these are refuted by this quarter. They are the reasons a 155% Rule of 40 score is not the end of the analysis.

What would change the view

At the Q3 report:

• Sequential US commercial growth. The year-over-year figure will stay large on base effects alone. Quarter-on-quarter is where deceleration shows up first.
• Remaining deal value conversion. Whether the $6.24 billion turns into recognised revenue on schedule, or the balance keeps growing because deals slip.
• Revenue against the $2.160–2.164 billion guide, and whether growth holds above 90% for a third consecutive quarter.
• Adjusted operating margin. Whether 62% survives the hiring that $6.24 billion of contracted work implies.
• Government concentration. Whether international commercial finally scales, or the US keeps carrying 81% of the company.
• Anything narrowing the 18-month visibility statement into something longer and more specific.

The read

Palantir delivered a quarter that is difficult to criticise on operating grounds: faster growth than last quarter at greater scale, margins most software companies never reach, cash generation funding all of it, and a guidance raise big enough to reset the year.

The open question is not whether Palantir is a good business on these numbers. It is what growth rate is still true in 2029, and a single quarter, however otherworldly, does not settle that.

Which means the useful thing for a holder to track is not the beat. It is the second derivative, quarter by quarter, and whether contracted deal value keeps converting on time. Those are the two series that show whether the growth is durable.

---

Sources and method. Every company figure here was read from Palantir’s own filings, not from coverage of them. The income statement, balance sheet, margins, cash flow, contract values, deal counts, guidance and the Karp quotation come from Exhibit 99.1 to the Form 8-K filed 2026-08-03; revenue is $1.935 billion as filed, shown as $1.94 billion where rounded. Total commercial and government revenue, the customer counts and the 157% net dollar retention figure are not in the results release — they come from Palantir’s Q2 2026 Business Update deck. “No debt” is our reading of the filed balance sheet, whose total liabilities contain no borrowings — only payables, deferred revenue, customer deposits and lease obligations.

The guidance verdict compares the results only with Palantir’s own previously issued outlook (Form 8-K, Exhibit 99.1, filed 2026-05-04) and with the year-ago quarter; the outlook figures were read from that filing.

Two things are not company figures and are attributed to the named coverage instead: consensus estimates and the “eight consecutive beats” record. Revenue consensus of ~$1.80 billion and adjusted EPS of ~$0.35 are the LSEG estimates reported by CNBC; ~$1.81 billion is the revenue consensus in TradingKey’s preview, which is also the source of the “eight consecutive beats” record; the segment consensus figures (US government ~$731 million, US commercial ~$716 million) are FactSet consensus as reported by MarketScreener — all cited above. Where outlets disagreed — revenue consensus at ~$1.80 billion and ~$1.81 billion — both are given as a range rather than a single value. This piece makes no claim about the share-price reaction.

Internal consistency was checked against the filing: segment revenue sums to total, the six-month figures reconcile, and the stated margins and the 155% Rule of 40 score follow from the reported revenue and operating income. The interpretation of how long the acceleration can last is our own analysis, not a company statement, and is not a forecast.

Related: Microsoft Q4 FY2026, the other side of the enterprise-AI spending question, and Meta Q2 2026, where capital spending was set against growth.

Tradune shows what you actually hold across every account, and whether a position has beaten a plain index fund since you bought it. Educational analysis, not investment advice.