INSTITUTION PROFILE / QUANTUM COMPUTING
DARPA (Defense Advanced Research Projects Agency)
It tests quantum computer makers' claims of reaching useful scale, starting from skepticism, and reports the results to the U.S. government
DARPA is a research and development agency inside the U.S. Department of War (formerly the Department of Defense). It does not build quantum computers, and it does not set standards or regulations. Its main role in quantum computing is to fund quantum computer companies while acting as an independent third party that checks whether their machines can really reach useful scale. Its program for this is the Quantum Benchmarking Initiative (QBI), which tests "whether a quantum computer whose computational value exceeds its cost can be built by 2033". The companies taking part come from several countries, and the results of the checks are passed to the U.S. government. This profile looks at one question: how much influence DARPA actually has over quantum computing. Company stages and plans change quickly, so each is given with a check date (16 October 2026).
- DARPA in 30 seconds
- What kind of body it is: the four types of influence
- What it does for quantum computing
- How far its influence reaches (powers and geography)
- The QBI stages and the organizations involved
- Outlook
- Glossary, references and claim-to-source audit
1. DARPA in 30 seconds
DARPA's quantum effort is not a contest to pick winners. It is an evaluation that tests each company's path one by one, with doubt as the starting point. The QBI page says it is not a competition between performers and that DARPA wants to evaluate all viable approaches for which funding is available Sourced. The program manager at the start said: "Our opening position is skepticism" (DARPA article, July 2024) Sourced. The page also says DARPA will pass the results of its checks to other stakeholders across the U.S. government Sourced. So DARPA's influence comes less from the size of its funding than from the weight inside government of the judgment "this passed DARPA's validation". How that judgment is used in other agencies' funding or procurement decisions is something this article could not confirm Not yet confirmed.
DARPA defines a quantum computer as having reached utility scale when the value of the computation it delivers exceeds the cost of the machine Sourced. Getting qubits to work in a laboratory is one thing. Being worth the money to build is another. QBI tests the second, with 2033 as the target. A quantum computer has to combine very many error-prone qubits to correct errors, so the required scale is very large. For more, see our explainers on logical qubits and fault-tolerant quantum computing, and on the reality of industrial applications.
2. What kind of body it is: the four types of influence
Unlike companies, the "institutions" in quantum computing articles do not build quantum computers themselves. They fall into four types, depending on the route by which they influence the industry.
3. What it does for quantum computing
DARPA's work on quantum computing is concentrated almost entirely in QBI and its predecessor Sourced.
| Activity | Timing | What it involves | Main players |
|---|---|---|---|
| Quantum Benchmarking Initiative (QBI) | Began in July 2024 (announced 16 July 2024) | Tests "whether a useful-scale quantum computer (one whose computational value exceeds its cost) can be built by 2033". Three stages: Stage A (describe a concept; up to US$1 million), Stage B (present an R&D plan, its risks and the prototypes needed; up to US$15 million), Stage C (work with the government to verify that the machine can be built as designed and operate as intended; up to US$300 million) | DARPA's government evaluation team. Run by the Multi X Office |
| US2QC (the pilot that came before QBI) | 6 February 2025: Microsoft and PsiQuantum move to the final stage | Examined whether little-explored approaches could reach useful scale. The two companies moved to the "validation and co-design" stage, which has the same goal as QBI's Stage C | Microsoft, PsiQuantum |
| New call for QBI entrants (Stage A, "QBIT") | Published 9 March 2026 | Organizations not yet funded by QBI can join Stage A (about six months) with new approaches. The deadline was 30 September 2026 in the March announcement, and the program page as of 8 October shows 30 November 2026 (we could not find the reason for the change) | New applicants |
| Call for independent verification and validation (IV&V) | Published 17 November 2025. Deadline 30 December 2026 | A call for parties to join QBI's verification effort (titled QBI Topic Independent Verification and Validation) | Applicants (details are in the call document) |
| Partnerships with states | Illinois (2024), Maryland (28 April 2025), New Mexico (2 September 2025) | Uses state-funded quantum test facilities for QBI validation. Illinois: the "Quantum Proving Ground". Maryland: the "Capital Quantum Benchmarking Hub", based in a University of Maryland research center. New Mexico: the "Quantum Frontier Project". Depending on results, DARPA and the state agreed to contribute matching amounts of up to US$100 million each (Maryland) and up to US$60 million each (New Mexico) over four years. The agreements do not require specific funding levels | The states, universities, national laboratories |
| Coordination with DOE | July 2024 | A framework with the DOE Office of Science to plan and coordinate research, development, engineering and testing of quantum computing. It also draws on a team at the Air Force Research Laboratory (AFRL) in Rome, New York | DOE Office of Science, AFRL |
Stage A checks on paper and in oral presentations whether a concept holds together. Stage B is about a year spent examining the assumptions and risks in the R&D plan. Stage C is where the government actually tests the hardware and systems. The managing director of QBI describes Stage C as moving "from plans to proof", closely scrutinizing whether the technology performs the way models and roadmaps say it should Sourced. The judgment rests on values the government has tested, not on a company's own stated values. This fits the thinking in our explainers on how to read quantum advantage claims and on coherence and fidelity.
DARPA's parent body is named the Department of War as of 16 October 2026 Sourced. On 8 October 2026 the Department announced new quantum computing initiatives and programmatic updates totaling about US$350 million (title and summary confirmed from war.gov's release feed) Sourced. However, this article could not confirm in the original text how much of that goes to DARPA's QBI, or the breakdown, because the site refuses automated retrieval Not yet confirmed. QBI's own budget for FY2026 and FY2027 is also not in the materials we could check.
4. How far its influence reaches (powers and geography)
| Dimension | What DARPA has | What it does not have |
|---|---|---|
| Powers | Authority to fund quantum computer companies and decide whether each advances to the next stage. Its government evaluation team verifies claims, and it passes the results to other U.S. government stakeholders. Executive Order 14413 directs agencies to share information on quantum computing supply chains, such as that generated by QBI, with the Departments of War, Commerce and Energy and others | Power to set standards, regulations or cryptography requirements. Power to decide other agencies' funding or procurement. Picking a winner is not QBI's purpose |
| Geography | Programs funded by the U.S. government, but participants also come from abroad. Of the six Stage C companies, Diraq is headquartered in Sydney, Australia. The 2025 Stage A group also included Alice & Bob (U.S. and France) and Oxford Ionics (UK and U.S.) | Other governments' policies. Power to manage participating companies' businesses |
| Binding force | The terms of each contract (cooperation with validation, results at each stage) | Power to require private companies to develop or publish anything. Power to guarantee that its results become an international yardstick |
| Size | About 250 government staff, nearly 100 program managers and about 300 projects (FY2024). FY2024 appropriations of about US$4.064 billion (DARPA as a whole). A cap of up to US$300 million per company at Stage C | A total budget for QBI alone, and DARPA's budgets for FY2026 and FY2027, are not in the materials we could check |
5. The QBI stages and the organizations involved
| Organization | Relationship | What the primary sources confirm |
|---|---|---|
| Atom Computing, Diraq, IBM, IonQ | Moved to Stage C on 7 October 2026 | Advanced after the year-long Stage B. Their approaches are, in order, neutral atoms, silicon CMOS spin qubits, modular superconducting processors and trapped ions |
| Microsoft, PsiQuantum | In Stage C via US2QC | Selected by DARPA for the final stage in February 2025. Microsoft uses a compact superconducting topological qubit architecture and PsiQuantum uses silicon photonics |
| Quantinuum, QuEra Computing, Xanadu and others | Active in Stage B | The companies active in Stage B are Nord Quantique, Photonic Inc., Quantinuum, Quantum Motion, QuEra Computing, Silicon Quantum Computing and Xanadu (page as of 8 October 2026) |
| Google Quantum AI, Quandela | Active in Stage A | Quandela joined Stage A in May 2026. The first Stage A group in 2025 also included Rigetti Computing and Alice & Bob, which do not appear in the current list of active participants (per the page's note, a company that finishes a stage is not listed until a contract for the next stage is awarded. We could not confirm the current status of these two) |
| DOE | Coordination | Signed a cooperation framework with the DOE Office of Science in 2024. DOE separately announced the Q Competition in September 2026. See our DOE profile |
| NIST (Department of Commerce) | A separate track | Standards and letters of intent sit on the Commerce side. See our NIST profile |
6. Outlook
| What can be read from the sources | Supporting primary information | Type |
|---|---|---|
| Validation of the six Stage C companies continues, and more companies may move up to Stage C. Selections from the new Stage A call are also expected | DARPA's news item of 7 October 2026 says that companies submitting under the new call are anticipated to be selected, and that promotions from Stages A and B are also expected. No timing is given | Not yet confirmed |
| DARPA judges that it is more likely than not that at least one company will build an industrially useful quantum computer by 2033. Which approach and which company is open | Stated as QBI's estimate on the program page (as of 8 October 2026). The page also says that multiple, one, or even no participants may ultimately show a path | Not yet confirmed |
| QBI's validation results could become a reference for how the U.S. government regards quantum companies | QBI says it will pass results to U.S. government stakeholders, and Executive Order 14413 directs agencies to share supply chain information that includes QBI's. We could not confirm that other agencies use that information to allocate funding | Inference |
| Partnerships with states may keep growing, with state budgets supporting QBI's validation facilities | DARPA says it will partner with any state that has already launched and funded quantum work, and it has agreements with three states | Inference |
| QBI at DARPA and the Q Competition at DOE are separate funding and separate evaluations. We could not confirm whether they use a common yardstick | We found no statement in either side's public materials that a common yardstick is used | Not yet confirmed |
7. Glossary
- QBI
- Quantum Benchmarking Initiative. DARPA's program to test whether a utility-scale quantum computer can be built by 2033.
- Utility scale
- A scale at which the value of a quantum computer's computation exceeds the machine's cost.
- US2QC
- Underexplored Systems for Utility-Scale Quantum Computing. A pilot program that came before QBI.
- Stage A / B / C
- QBI's three stages: presenting a concept, examining the R&D plan, and verifying with the government that the machine can be built as designed.
- IV&V (independent verification and validation)
- A third party, independent of the developer, checking performance claims.
- Program manager
- At DARPA, the person who designs and runs a research program. Terms are limited in length.
8. References
- DARPA About DARPA (its status as an independent R&D agency within the Department of War, its mission, its director; checked 16 October 2026). https://www.darpa.mil/about
- DARPA DARPA FY 2024 Agency Financial Report (established 1958, about 250 government staff, nearly 100 program managers, about 300 projects, FY2024 appropriations). https://www.darpa.mil/sites/default/files/attachment/2024-11/darpa-2024-afr-final.pdf
- DARPA Quantum Benchmarking Initiative (QBI's goal, the three stages and their caps, active companies, calls; as of 8 October 2026). https://www.darpa.mil/research/programs/quantum-benchmarking-initiative
- DARPA Multi X Office (the office that runs QBI; an expansion of the former Microsystems Technology Office). https://www.darpa.mil/about/offices/mxo
- DARPA Moving quantum computing from hype to prototype (16 July 2024; announcement of QBI, the "skepticism" remark, coordination with the DOE Office of Science, Illinois and AFRL). https://www.darpa.mil/news/2024/quantum-computing-prototype
- DARPA DARPA selects two discrete utility-scale quantum computing approaches for evaluation (6 February 2025; Microsoft and PsiQuantum, US2QC). https://www.darpa.mil/news/2025/quantum-computing-approaches
- DARPA DARPA eyes companies targeting industrially useful quantum computers (3 April 2025; the first Stage A group). https://www.darpa.mil/news/2025/companies-targeting-quantum-computers
- DARPA DARPA, State of Maryland sign agreement to propel quantum research (28 April 2025; the Capital Quantum Benchmarking Hub, up to US$100 million each). https://www.darpa.mil/news/2025/agreement-propel-quantum-research
- DARPA DARPA, State of New Mexico establish framework to advance quantum computing (2 September 2025; the Quantum Frontier Project, up to US$60 million each). https://www.darpa.mil/news/2025/darpa-new-mexico-establish-framework-advance-quantum-computing
- DARPA Quantum Benchmarking Initiative expands quest to separate hype from reality (10 March 2026; new Stage A call, eleven Stage B organizations, deadlines). https://www.darpa.mil/news/2026/qbi-stage-a-qbit
- DARPA Four more teams enter Quantum Benchmarking Initiative's final stage (7 October 2026; four Stage C companies, Quandela joining, expected promotions). https://www.darpa.mil/news/2026/four-more-teams-enter-quantum-benchmarking-initiative-final-stage
- Department of War Department of War Announces $350 Million in Quantum Computing Initiatives (8 October 2026; title, summary and date confirmed from war.gov's release feed; the full text could not be retrieved because the site refuses automated access). https://www.war.gov/News/Releases/Release/Article/4622109/department-of-war-announces-350-million-in-quantum-computing-initiatives/
- The White House Executive Order 14413, Ushering in the Next Frontier of Quantum Innovation (22 June 2026; Section 6(c) on sharing QBI information). https://www.whitehouse.gov/presidential-actions/2026/06/ushering-in-the-next-frontier-of-quantum-innovation/
9. Claim-to-source audit
| Claim in the text | Type | Source |
|---|---|---|
| An independent R&D agency within the Department of War, set up after Sputnik, the parent's name as of 16 October 2026 | Sourced | Reference 1 https://www.darpa.mil/about |
| Established 1958, about 250 government staff, nearly 100 program managers, about 300 projects, FY2024 appropriations of about US$4.064 billion | Sourced | Reference 2 https://www.darpa.mil/sites/default/files/attachment/2024-11/darpa-2024-afr-final.pdf |
| QBI's goal, the three stages and caps, "not a competition", the definition of utility scale, active companies per stage, the 2033 estimate, call deadlines (30 November and 30 December 2026) | Sourced | Reference 3 https://www.darpa.mil/research/programs/quantum-benchmarking-initiative |
| QBI is run by the Multi X Office | Sourced | Reference 4 https://www.darpa.mil/about/offices/mxo |
| The announcement of QBI (July 2024), the "skepticism" remark, coordination with the DOE Office of Science, Illinois and AFRL | Sourced | Reference 5 https://www.darpa.mil/news/2024/quantum-computing-prototype |
| Microsoft and PsiQuantum in the final US2QC stage (6 February 2025) | Sourced | Reference 6 https://www.darpa.mil/news/2025/quantum-computing-approaches |
| The 2025 Stage A group (including Alice & Bob, Oxford Ionics and Rigetti Computing) | Sourced | Reference 7 https://www.darpa.mil/news/2025/companies-targeting-quantum-computers |
| Partnership with Maryland (28 April 2025; up to US$100 million each over four years) | Sourced | Reference 8 https://www.darpa.mil/news/2025/agreement-propel-quantum-research |
| Partnership with New Mexico (2 September 2025; up to US$60 million each over four years); Sandia and Los Alamos on the validation team | Sourced | Reference 9 https://www.darpa.mil/news/2025/darpa-new-mexico-establish-framework-advance-quantum-computing |
| The new Stage A call (March 2026), eleven Stage B organizations, the 30 September deadline | Sourced | Reference 10 https://www.darpa.mil/news/2026/qbi-stage-a-qbit |
| The four Stage C companies (7 October 2026), Quandela joining, expected promotions, the description of Stage C | Sourced | Reference 11 https://www.darpa.mil/news/2026/four-more-teams-enter-quantum-benchmarking-initiative-final-stage |
| The Department of War's announcement of about US$350 million in quantum initiatives on 8 October 2026 | Sourced | Reference 12 https://www.war.gov/News/Releases/Release/Article/4622109/department-of-war-announces-350-million-in-quantum-computing-initiatives/ |
| Executive Order 14413's direction to share QBI information | Sourced | Reference 13 https://www.whitehouse.gov/presidential-actions/2026/06/ushering-in-the-next-frontier-of-quantum-innovation/ |
| The four-type grouping of institutions | Our calculation | This article's own grouping |
| The QBI share and breakdown of the US$350 million, DARPA's FY2026 and FY2027 budgets, use of QBI results in other agencies' funding, a common yardstick with DOE, the reason for the changed Stage A deadline | Not yet confirmed | Not in References 3, 10 or 12, or the original text could not be checked. This article does not estimate them |
| More Stage C promotions, and whether a utility-scale machine appears by 2033 | Not yet confirmed | Reference 3 https://www.darpa.mil/research/programs/quantum-benchmarking-initiative and Reference 11 describe these as expectations and estimates, not settled facts |
| The two items in the outlook in Section 6 (QBI results as a government reference, growth of state partnerships) | Inference | This article's reading of References 3, 5, 8, 9 and 13 |
Last updated 16 October 2026 / Troy Technical
Figures and names come from DARPA's website and news items, its financial report, the Department of War's release list and a U.S. executive order. Company stages and plans change quickly, so the check date (16 October 2026) is stated throughout. Items marked "Inference" are this article's reading of the primary information, not statements by DARPA.
