On August 3, 2026, Palantir reported second-quarter results that pushed the company past a milestone few enterprise software vendors reach: revenue growing faster the larger the base gets. Quarterly revenue reached $1.935 billion, up 93% year over year and 19% sequentially, and management raised full-year guidance for the third quarter in a row. Shares, already among the most expensive in large-cap software, moved sharply higher after the print.
The report is worth reading closely not because the headline growth rate is surprising, but because of where the growth is now coming from. For most of Palantir's public life the story was government contracts. This quarter it was the commercial business, and specifically the US commercial business, that did the heavy lifting.
The commercial engine took over
US commercial revenue reached $764 million, up 149% year over year and 28% sequentially. Total contract value bookings in that same US commercial segment rose 153% to $2.132 billion. When bookings grow faster than recognized revenue, it usually means the pipeline being signed today is larger than the revenue being reported today, which is the pattern investors look for when they want to believe a growth rate is durable rather than a one-quarter spike.
The deal-count disclosure filled in the texture. Palantir said it closed 220 deals worth at least $1 million, including 98 worth at least $5 million and 70 worth at least $10 million. Those are not pilots. A company does not sign 70 eight-figure contracts in a single quarter on the strength of a demo; it signs them because earlier deployments produced something the buyer could measure.
Bookings grew faster than revenue in Q2 2026
Palantir year-over-year growth by measure. When bookings outrun recognized revenue, the pipeline being signed is larger than the revenue being booked.
Total Q2 2026 revenue was $1.935 billion. US commercial revenue was $764 million and US commercial total contract value bookings were $2.132 billion.
The mechanism Palantir keeps pointing to is its Artificial Intelligence Platform, or AIP, and the "bootcamp" motion that puts it in front of prospective customers. The company described one competitive win in which a large Silicon Valley technology firm ran a bake-off, chose Palantir over a frontier AI lab's own tooling, and converted to a $10 million annual contract. That anecdote matters beyond the dollar figure, because it speaks to a question the whole industry is now circling: once every company can call the same underlying models, what is the defensible layer?
Why a raw model is not the same as a deployed one
Palantir's answer, implicit in these results, is that the defensible layer is the orchestration and data plumbing that sits above the model. A frontier model can draft, summarize and reason, but it cannot on its own reconcile a manufacturer's inventory system with its logistics data, enforce who is allowed to see which record, and write the result back into an operational workflow that a factory floor actually trusts. That connective tissue is what enterprises are paying for, and it is largely model-independent by design.

This is the same structural bet that model-agnostic platforms across the market are making, including tools like Metir that route work across many providers rather than committing to one. The value migrates to whoever owns the workflow, the context and the permissions, because those do not become obsolete when a new model ships the following month. Palantir's commercial acceleration is one large, public data point in favor of that thesis, though it is not the only possible reading.
The large deals are the leading indicator
The clearest forward signal in the quarter was the concentration of very large contracts. Enterprise AI has spent two years accused of a "pilot to production" gap, in which experiments impress but rarely convert into budgeted, recurring spend. A quarter with 70 contracts of at least $10 million is evidence that at least some buyers have crossed that gap.
Seventy eight-figure deals in a single quarter
Palantir deals closed in Q2 2026 by minimum contract size. Large-deal concentration is the clearest sign that buyers have crossed the pilot-to-production gap.
The $1M, $5M and $10M tiers are cumulative thresholds, not exclusive buckets. Palantir raised full-year 2026 revenue guidance to about $8.154 billion after the quarter.
A company does not sign 70 eight-figure contracts in a single quarter on the strength of a demo.
On the Q2 2026 deal mix
Palantir raised its full-year 2026 revenue guidance to roughly $8.154 billion, which would represent about 80% annual growth, and lifted its US commercial revenue outlook to more than $3.42 billion, up from prior guidance of $3.22 billion. It also reported a record Rule of 40 score of 155% and $1.22 billion in adjusted free cash flow, meaning the growth is not being purchased with cash-burning losses. On the standard software scorecard, this is close to a best-case quarter.
The part the numbers cannot settle
None of this resolves the valuation question, and any neutral read has to name it. Palantir trades at one of the richest price-to-sales multiples in large-cap software, which means the market is already pricing in years of continued high growth. When a stock is valued that way, even excellent results can be followed by volatility, because the bar was set by expectations rather than by fundamentals. Two things can be true at once: the business is executing at a very high level, and the shares carry more risk than the operating numbers alone suggest.
There is also a durability question the quarter cannot answer. US commercial growth of 149% is partly a function of a smaller base and a specific selling motion that is working right now. The honest position is that a bookings figure growing faster than revenue is a genuinely encouraging sign, but one quarter does not prove a multi-year trajectory. The next two reports, and whether the international commercial segment starts to accelerate the way the US segment has, will say more.
Anthropic
NVIDIAWhat it signals for the AI market
Strip away the ticker and Palantir's quarter is a reading on a broader question: is enterprise AI actually converting into revenue, or is it still mostly experimentation funded by innovation budgets? This one company, in this one quarter, says conversion is real for buyers who have a platform to operationalize it. That does not generalize to every vendor, and it does not make the shares cheap. It does suggest that the market is beginning to reward the layer that turns a capable model into a deployed system, rather than the model in isolation.
For anyone building or buying AI tooling, the takeaway is less about Palantir and more about where value is accruing. The model is increasingly a commodity input. The orchestration, the data access controls and the workflow that make it useful are not. Palantir's Q2 is a large, audited example of that gap being paid for in cash.
Sources:
- Palantir (PLTR) earnings Q2 2026, CNBC
- Palantir stock rises on soaring commercial revenue and AI sovereign tools, CNBC
- Palantir Q2 2026 earnings call highlights, GuruFocus
- Palantir Technologies Q2 2026 press release, SEC EDGAR
- Palantir Q2 2026: revenue up 93%, US commercial 149%, TradingKey
Image credits
- Hero: "Alex Karp attends AI Safety Summit" by UK Government, licensed CC BY 2.0, via Wikimedia Commons. Retrieved August 6, 2026, and visually reviewed before use. The image shows Palantir CEO Alex Karp; it is used to illustrate the company and predates the Q2 2026 results discussed here.