DOE's $215M Quantum Q Competition Has a $250,000 Entry Ticket and a $200M Ceiling. Read the Instrument Before the Number.

The single most revealing number in the Department of Energy's new quantum competition is not $215 million. It is $250,000 — the award floor DOE listed on grants.gov, and the exact amount a participating company receives when DOE approves its technical and validation plan. The second most revealing number is $2.5 million: everything DOE has actually appropriated for this in fiscal 2026.

Everything between those two figures is a promise contingent on future acts of Congress.

DE-FOA-0003657, The DOE Quantum Genesis Q Competition, posted September 17, 2026 under the Office of Science (Assistance Listing 81.049). Applications are due October 19, 2026, with an archive date of January 19, 2027. DOE lists approximately 10 expected awards, an award floor of $250,000, an award ceiling of $200,000,000, total planned funding of $215,000,000, and no cost-share requirement. A virtual informational webinar is scheduled for September 25, 2026 at 1:00 p.m. Eastern. Questions go to qcompetition@doe.gov.

Note the instrument field: Other. This is not a grant and not a cooperative agreement. DOE posted a sample Other Transaction Agreement alongside the announcement, and that document — not the funding total — is the thing to read first.

What the competition is actually buying

The Quantum Genesis initiative was announced June 23, 2026, a day after two quantum executive orders were signed, including Ushering in the Next Frontier of Quantum Innovation and its Quantum Computer for Application Development and Discovery Science (QC-ADDS) effort. Quantum Genesis sits underneath the broader Genesis Mission, which we covered in DOE's Genesis Mission: $293 Million for AI and Quantum Research.

The stated target is a scientifically relevant quantum computer (SRQC): at least 100 logical qubits capable of hundreds of millions of fault-tolerant operations, plus a scientific demonstration program proving the machine does real work in chemistry, materials, physics, or applied mathematics. DOE's quantum information science lead, Tanner Crowder, framed it plainly on the announcement call: "Our goal is not to have a bunch of qubits that can't do much."

The payout structure is two-stage, and it is not symmetric.

Phase I pays fixed milestone awards of up to $1.5 million per participant, split into $250,000 on DOE approval of the technical and validation plan and $1.25 million on delivery of a validated intermediate milestone — a prototype device.

Phase II is where the money is. A $100 million general incentive pool gets divided among participants who demonstrate a first-generation SRQC at the 100-logical-qubit threshold. Two additional $50 million bonus pools are reserved for machines that reach 150 and 200 logical qubits respectively.

Add those up: $100M + $50M + $50M = $200 million, which is exactly the award ceiling DOE published. The ceiling is not a drafting artifact. It is the arithmetic of a single participant sweeping every Phase II pool.

Why the instrument matters more than the headline

An Other Transaction Agreement is not subject to the Uniform Guidance at 2 CFR Part 200. That single fact reshapes the calculus for anyone weighing whether to compete.

What you gain: no federally negotiated indirect cost rate, no standard grant audit architecture, and negotiable intellectual property terms rather than default Bayh-Dole allocation. For a venture-backed hardware company whose board treats IP as the balance sheet, that flexibility is the reason to be in the room at all.

What you give up: the procedural protections that come with assistance awards. No standard dispute or appeal path. Termination terms are whatever the agreement says. The milestone definitions — "validated intermediate milestone," "first-generation SRQC" — are defined in the agreement and validated by DOE, not by a peer review panel working from published criteria.

NSF has been running the same play with X-Labs, its $1.5 billion milestone-driven initiative built outside the standard grant architecture. Two agencies independently reaching for OTAs on their highest-visibility programs is a trend line, not a coincidence. Federal capital for frontier technology is migrating out of grant instruments — a pattern we also traced in the Office of Strategic Capital's $500 million to $1 billion credit facility program.

The eligibility screen is narrow, and it is deliberate

Grants.gov lists exactly two applicant types: small businesses and for-profit organizations other than small businesses. Universities cannot apply. National laboratories cannot apply. Nonprofits cannot apply.

That is not an oversight — DOE built a parallel track for the labs. A separate Quantum High-Performance Computing Validation and Verification Testbed lab call carries $45 million in total planned funding, $14 million of it in FY 2026, and is open only to DOE National Laboratories. The testbed's job is to build hardware-agnostic tools to characterize and benchmark whatever the companies deliver.

Read the two announcements together and the architecture is clear: industry builds the machine, the labs referee the claims, and DOE writes the check on validated results. If you are an academic quantum group, your path into Quantum Genesis runs through a lab partnership or a company's scientific demonstration program — not through this solicitation. Existing NQIS Research Centers and the DOE quantum center renewals remain the academic lane.

Can anyone actually win Phase II?

DOE officials were notably candid about the odds. Under Secretary for Science Darío Gil said 100 logical qubits is possible "as early as 2028" — a forecast, not a commitment — and described the moment as "50 years in the making."

The public record supports the caution. Quantinuum has shipped systems with roughly 50 error-corrected logical qubits. IBM and the University of Chicago reported a July 2026 experiment involving 70 logical qubits on a task IBM argues is beyond practical classical simulation. Infleqtion has demonstrated 12. Google's Willow chip established below-threshold error correction in late 2024.

So the 100-qubit base threshold sits perhaps a factor of 1.5 above the current published frontier, and the 200-qubit bonus tier roughly 3x. The "hundreds of millions of fault-tolerant operations" requirement may be the harder constraint — logical qubit counts get headlines, sustained fault-tolerant circuit depth gets papers.

The honest read: Phase II is a prize, and prizes are priced for the possibility that nobody collects. DOE wrote the announcement to say so, noting explicitly that failure to participate or complete the milestones does not disqualify an entity from future DOE collaborations or procurements. That is unusually gracious language for a federal solicitation, and it tells you DOE knows some strong candidates have roadmaps that do not line up with this window.

What to weigh before October 19

Price the Phase I ticket honestly. For a company already on a fault-tolerance roadmap, $1.5 million across two milestones is real but not transformative — a fraction of one quarter's burn at a serious hardware company. The durable value in Phase I is structural: DOE approval of your technical and validation plan, standing access to National Laboratory and NQIS Research Center expertise, and a validated third-party benchmark you can put in front of investors and customers. Several firms will find that worth more than the cash.

Do not build a plan around Phase II cash flow. With $2.5 million appropriated in FY 2026 and everything else contingent on outyear appropriations, the incentive pools are a policy intention. The House Appropriations Committee has already signaled it will not fund additional AI supercomputers at DOE without a better understanding of infrastructure, operations, and outyear funding implications for the Genesis Mission, and it requested a briefing on expected future support levels. A company that treats $200 million as bankable is misreading both the instrument and the appropriations posture.

Send a lawyer to the sample OTA before you send an engineer to the webinar. Milestone acceptance criteria, validation authority, IP allocation, and termination terms are all negotiable in an OT — which means they are all losable. The September 25 webinar is the right venue to ask which terms DOE considers fixed and which are open.

Watch the validation testbed. Whoever builds the benchmarking stack at the national labs effectively defines what "100 logical qubits" means for payout purposes. If your architecture is unusual — neutral atom, photonic, topological — the definitions written into that $45 million lab call will matter more to you than the size of the bonus pool.

DOE's own framing is that the Q Competition is a first step toward machines it hopes to deploy at National Laboratories into the 2030s. Applicants should price it the same way: the opening round of a long relationship with a customer, in which the prize money is upside rather than the business case. For teams sizing up a first federal pursuit of this kind, Granted can help translate a technical roadmap into the structured milestone language these announcements are now written in.