Oklo's quarter: the beat was a land sale, and the market priced the reactors at nothing.

Oklo Inc. (NYSE: OKLO) filed its third-quarter Form 10-Q on November 6, 2026, and the stock rose 22.7%. In the quarter it reported, the company sold no electricity, no heat and no isotopes. The headlines said the loss was narrower than expected and revenue beat the estimate. Both statements are true. Neither means what it sounds like. This Comuvia News analysis reads the filing line by line against the company's own update deck and the earnings call, and says which claims the documents actually support.

Where the beat came from. The reported loss was $(0.22) per share against a consensus near $(0.27). Inside it sits an $11,430 thousand gain on sale of asset. The filing's notes are explicit about what it was: "the Company recognized a gain of $11,430 in the condensed consolidated statement of operations related to the sale of land to a third party." No comparable item appears in any earlier period shown. Add the land back, because selling land is not something a reactor company can repeat next quarter, and the loss becomes $53.2M, or $(0.28) per weighted-average share. On the reported basis this was a five-cent beat; without the land sale it is about a cent worse than expected. We put our confidence that the gain is genuinely one-off at about 95%, because the filing names a single transaction, and our confidence that this makes it a miss on the basis analysts actually used at about 55%, because we could not establish whether their estimate already excluded such gains.

Where the revenue came from. All $1,861 thousand of it is attributable to the two businesses Oklo bought in June 2026 - ARMEC, a precision manufacturer, and CEI, which brings about thirty years of sodium-systems experience. The filing states that "revenue attributable to the acquisitions ... was $1,861" and that those same acquisitions produced a net loss of $1,114 thousand in the quarter. The revenue disaggregation splits the line into $1,561 thousand of engineering and consulting services and $300 thousand of manufacturing and fabrication services, so it is primarily, not wholly, consulting work. Buying the two companies was a defensible decision and the deck is specific that both are already working on Aurora specifications and supplier reviews. But understand what the revenue beat measured: machine-shop and consulting work at two companies acquired four months into the year. It told you nothing about reactors, fuel or isotopes.

What the market was paying for the reactors. With $3,130.8M of cash and marketable securities and 195,289,593 shares outstanding on the filing's cover, Thursday's $19.20 close implied about $0.6B for everything Oklo has built - roughly 16% of the company's value, with the other 84% being the money in the bank. Friday's $23.55 close implied about $1.5B. Nothing physical changed between the two closes, and the move was sector-wide: NuScale rose about 15% and Nano Nuclear about 9% in the same session without reporting anything. Separately, Oklo sold 32,608,356 shares under its at-the-market programmes in the nine months to September at an average net price of $70.19 - about three times Friday's close.

Contracted, or narrated. The filing states that Oklo had no material commitments or contractual obligations as of September 30, 2026. The same week's deck leads with a 27-module manufacturing reservation, a 12 GW agreement with Switch and the 1.2 GW Meta campus. Both are true. The gap between them is the point, and it is the single most useful thing a reader can carry away from this quarter.

And the other half of the ledger. Groves, in Lockhart, Texas, went from an open field to first criticality in under a year - 229 days of substantial construction, about $65M of capital plus about $10M of fuel, critical on August 5, 2026. Aurora-ID has its preliminary safety analysis approved and structural steel going up. The fuel fabrication facility has four Department of Energy safety approvals and equipment going in. That is a real deployment record and the analysis says so. The constraint that decides whether it scales is fuel, not reactors: an independent read on the record puts around $3,000 per kilogram of uranium for 5% enriched oxide fuel against roughly $15,000 per kilogram at 20% enrichment, before any fabrication at all.

On the record (original audio, attributed on screen):

  • Oklo (company channel) - inside the Groves reactor reaching first criticality.
  • U.S. Department of Energy (official channel, public domain) - Greg Beard, head of the DOE loan office, at the Perry plant, Ohio, October 5, 2026, announcing a $4.2 billion conditional commitment to Vistra.
  • Decouple (programme channel) - what the West's first small modular reactor actually costs.
  • Oklo (company channel) - surplus plutonium as a bridge fuel.
  • Decouple (programme channel) - why fuel, not reactors, is the hard industrial problem.

What we checked in the documents. The deck's financial appendix contains two reproduction errors; the filing is correct in both cases. Its balance-sheet footnote prints the prior year's share counts and dates, understating shares by about 18%, and one column is headed "Nine Months Ended December 31". Separately, two major market-data sites were carrying a stale 186.02M share count against the filing cover's 195,289,593 - for this company, compute market value from the filing cover, not from a data vendor.

Every figure above is located in a retained copy of its source, and the record page lists each one with its publisher, its capture date and the digest of the bytes quoted. Where a transcript on screen differs from the retained caption track or recording, the record page says so and gives both.

This is research and analysis, not investment advice. No position is recommended and nothing here is a forecast of any share price.

Comuvia News. Evidence from the record.