Two names, two different things
NSTXL is an organization. S2MARTS is a contract vehicle. The organization did not create the vehicle and does not decide who wins under it; a Navy command did, and does. Once that distinction is clear, most of the confusion around consortium contracting resolves itself, because the two entities answer different questions. NSTXL answers how do I get in the room. The government sponsor answers do I want your solution. Firms that conflate the two write a membership check and then wait for something that was never being sold.
The larger frame matters too. Other transactions are not contracts under the Federal Acquisition Regulation. The Department of Defense carries out prototype projects under 10 U.S.C. § 4022 and research projects under § 4021, and neither instrument is a procurement contract, so the FAR does not apply on its own. Use of the authority has grown sharply: the Congressional Research Service reports that the number of DoD other transaction actions rose from 496 in FY2017 to 4,391 in FY2022. A large share of that volume moves through consortia, and consortia are run by management firms. NSTXL is one of the two or three that matter most.
What NSTXL is
The National Security Technology Accelerator is a nonprofit based in Arlington, Virginia, founded and led by Tim Greeff. It does not buy anything. It holds a base other transaction agreement with a government sponsor, recruits and vets a membership, distributes the sponsor's requests to that membership, collects responses, and administers the resulting agreements and payments.
Its own FAQ is unusually direct about the boundary. NSTXL "receives proposed solutions and provides to Government for evaluation," and states plainly that it "is not involved in the selection process or final decision." Read that as the operating rule for every conversation you have with consortium staff. They can tell you what a request means, what format the government expects, and when the window closes. They cannot tell you whether you will win, and they do not score anything.
The money runs the way you would expect once you know the structure. Per NSTXL's published answer, "NSTXL is paid a management fee by the Government which is a percentage of the total Project Award," and that fee "does not affect the nominal value of the award to the Performer." So the manager's fee is not carved out of your price. It still sits inside the sponsor's total, which is worth knowing when you are reading a request's stated funding ceiling.
As of September 2025, MITRE's acquisition reference lists three NSTXL-managed vehicles: S2MARTS, S2MARTS Research, and the Space Enterprise Consortium. The Army's Training and Readiness Accelerator, which NSTXL once ran, moved to Advanced Technology International for its second iteration under Army PEO STRI. Consortium management is itself competed, and manager turnover is a real feature of this market.
What S2MARTS is
S2MARTS stands for Strategic & Spectrum Missions Advanced Resilient Trusted Systems. It is sponsored by Naval Surface Warfare Center, Crane Division, and NSTXL announced the award on March 12, 2019, describing the agreement as one that "will be active for up to 10 years and does not have a finite ceiling." That last clause is the structural point: unlike an IDIQ with a stated maximum, this vehicle's capacity is set by the appropriations the sponsor and its partner offices bring to it, not by a number written into the base agreement.
The subject matter is narrow and specific. S2MARTS describes itself as the department's rapid contracting vehicle for trusted microelectronics, strategic and spectrum missions, and adjacent critical mission areas. The named focus areas are radiation-hardened microelectronics, RF and optoelectronic microelectronics, new microelectronics development including system-on-chip co-development, design assurance, and verification and validation methods for microelectronics test. If your capability does not touch one of those, the vehicle is probably not your lane no matter how good the technology is.
Scale, as reported by the consortium manager in March 2025: roughly $3 billion awarded across 54 opportunities to 133 performers, drawn from a membership of almost 1,000 organizations of which more than 81 percent were nontraditional defense contractors. Those are the manager's own figures and should be read as such, but the shape they describe is consistent with the award record. Separately, NSTXL's consortium page reports more than $5 billion obligated across roughly 200 prototype projects across the vehicles it runs.
One number needs care. NSTXL advertises a network of more than 18,000 "innovators." That is a count of registered individuals, not companies. The company-level figure for S2MARTS is the roughly 1,000 member organizations reported in 2025. Treating the larger number as a competitor count will give you a wrong picture of the field in both directions.
Two statutes under one roof
S2MARTS started with prototype authority under 10 U.S.C. § 4022, formerly § 2371b. In October 2021 it added research authority under § 4021, formerly § 2371, which lets the same vehicle carry basic, applied, and advanced research projects. NSTXL's stated reason for wanting both was continuity: the addition "gives our clients the unique ability to advance from research to production all under one OTA vehicle."
That produced S2MARTS Research as a distinct lane with its own membership terms. Organizations that want to participate only in research projects can join at no cost, and existing NSTXL members reach those opportunities "without additional fees or membership requirements." For a university group or a research-stage company, the free research membership is the cheapest legitimate way to see how this vehicle operates from the inside.
The gate that has to open
A prototype other transaction is only available when one of four conditions at 10 U.S.C. § 4022(d)(1) is satisfied. At least one nontraditional defense contractor or nonprofit research institution participates to a significant extent. Or every significant non-federal participant is a small business or a nontraditional defense contractor. Or at least one third of total project cost comes from non-federal sources. Or the agency's senior procurement executive determines in writing that exceptional circumstances justify the arrangement.
The term "nontraditional defense contractor" is defined at 10 U.S.C. § 3014, and the test is Cost Accounting Standards coverage, not company size or revenue. An entity qualifies if it is not currently performing, and has not performed for at least one year before the solicitation, any DoD contract or subcontract subject to full CAS coverage. NSTXL's own FAQ makes the practical point: most entities qualify, because small business exemptions and the CAS dollar thresholds keep the great majority of firms outside full coverage.
This is why the 81 percent nontraditional figure is not a marketing statistic. It is the mechanism by which the gate stays open. A team built entirely of large CAS-covered primes has a harder qualification story than a team with real nontraditional content, and sponsors know it.
How closely a capability matches the S2MARTS subject matter
Our editorial weighting of subject-matter fit, read from the vehicle's published focus areas and its public award record. Illustrative rather than measured, and it says nothing about any individual firm's chance of winning.
Two of those rows deserve a sentence of explanation. Software scores well when it is embedded in a named system the sponsor is already buying, and poorly when it stands alone, because the vehicle's stated purpose is the platforms and components themselves. A back-office analytics product is not a bad product; it is a mismatch with this particular front door, and there are better ones.
How a project actually moves
S2MARTS solicits through what it calls a Request for Solutions. The Space Enterprise Consortium calls its equivalent a Request for Prototype Proposals. The names differ; the shape does not. The government writes the requirement, the manager pushes it to members, members respond through the manager's portal, the government evaluates, and the manager papers the award.
Membership is a hard prerequisite. NSTXL states it without qualification: "You must be a member of NSTXL to submit a response to any request." There is no path where a non-member submits and sorts out membership later, which is the single most common reason a firm watches a well-fitting request close without bidding.
From membership to signed agreement
End-to-end timing spreads wide, and the published figures are best read as ends of a range rather than an expectation. NSTXL advertises award in as little as 70 days and roughly 40 percent faster time to award than the alternative. SpEC material describes a Request for Prototype Proposals reaching award in as little as 85 days. Against that, the consortium's own case study for the Space Force ground command-and-control effort describes request to award in under eight months for a large, integrated program with 15 nontraditional sub-performers. Small and tightly scoped moves fast. Large and integrated does not.
Payment mechanics are worth two lines because they change working capital planning. Milestones are the unit of payment, not incurred cost, and NSTXL states that it pays within 30 calendar days of receipt. A firm structuring a bid should look hard at whether the first payable milestone lands early enough to fund the second.
What membership costs
Dues are published and modest, and one fee covers every NSTXL-managed vehicle. The tiers are set by entity type and revenue.
| Entity type | Band | Annual dues |
|---|---|---|
| Corporate | Under $10 million revenue | $250 |
| Corporate | $10–50 million / $50–100 million / above $100 million | $1,000 / $5,000 / $10,000 |
| Non-profit | Under $50 million / $50–100 million / above $100 million | $500 / $5,000 / $10,000 |
| University or laboratory | Flat | $2,500 |
| Incubator | Flat | $500 |
| Research-only participation | S2MARTS Research lane | No cost |
For a firm under $10 million in revenue, $250 a year is not a decision that needs a business case. The real cost of this lane is never the dues. It is the engineering and writing time a serious response consumes, and the opportunity cost of the requests you chase that were shaped around somebody else's existing work.
The three vehicles side by side
| Axis | S2MARTS | S2MARTS Research | SpEC |
|---|---|---|---|
| Government sponsor | NSWC Crane Division | NSWC Crane Division | U.S. Space Force Space Systems Command |
| Statutory authority | 10 U.S.C. § 4022, prototype projects | 10 U.S.C. § 4021, basic, applied and advanced research | 10 U.S.C. § 4022, prototype projects |
| Subject matter | Trusted and rad-hard microelectronics, spectrum and strategic missions, hypersonics | Research feeding the same technical areas, including materials work | Space systems, ground segment, command and control, launch-adjacent technology |
| How work is solicited | Request for Solutions to the membership | Research solicitations to the membership | Request for Prototype Proposals to the membership |
| Membership | Tiered dues, one fee covers all NSTXL vehicles | Free for research-only participants | Same NSTXL membership |
| Extra registration step | CAGE code and UEI | CAGE code and UEI | CAGE code, UEI, and a DD-2345 at registration |
SpEC is the other vehicle worth understanding, because the same membership reaches it. Space Systems Command created the consortium in 2017 and selected NSTXL as manager in 2020, with the ceiling raised to $12 billion over a ten-year term. The DD-2345 requirement at registration is the operational detail that trips firms up: that is the Militarily Critical Technical Data Agreement, it is certified through the Defense Logistics Agency's Joint Certification Program, and getting it takes real calendar time. Start it before you need it.
Dues buy standing, not a pipeline
Membership makes a firm eligible to respond and puts it on the distribution list. It does not create a requirement, does not influence evaluation, and does not give the manager any ability to advocate for a member. Against roughly 1,000 member organizations on one vehicle alone, standing is the floor, not the advantage. The advantage is a sponsor who already understands what you build.
Intellectual property under these agreements
Because an other transaction is not a procurement contract, the DFARS data rights clauses do not attach on their own. Everything is negotiated. NSTXL states the rule for its vehicles plainly: "IP and Data Rights are negotiated on a project-by-project basis prior to the final agreement," and commercial intellectual property developed before the engagement remains the sole property of the vendor.
The word doing the work in that second clause is "before." A firm that can show, with dates and artifacts, what existed prior to the project keeps it. A firm that cannot has an argument, not a position. Practical consequences follow.
Write the background IP schedule before the negotiation, not during it. List every model, library, dataset, and design file that predates the project, with a date you can evidence from a repository or a release record.
Grant licenses per deliverable, not per project. Government purpose rights in an integration layer alongside narrower rights in a core component is a normal outcome. A single project-wide grant is not, and it is what an opening template often proposes.
Name third-party and open-source components with their licenses. In a microelectronics or spectrum program this includes toolchains, reference designs, and vendor IP blocks, which carry their own restrictions that no consortium agreement can override.
Treat data as separate from software. Test data, characterization results, measured performance, and trained model weights each have independent value, and a clause written for "technical data" often fails to say which of them it reaches.
Who has been winning
The public record for 2025 and 2026 shows a vehicle running at scale and skewing toward large integrated efforts. In January 2025 the Department awarded Kratos the Task Area 1 systems engineering, integration and test role on MACH-TB 2.0, an other transaction agreement with a five-year period and a total value of $1.45 billion if all options are exercised; S2MARTS presents the MACH-TB program lineage as its flagship case study with more than $2 billion in obligated funding. In January 2026 the office of the Under Secretary for Research and Engineering awarded the PROTEAN electromagnetic warfare and seeker prototyping effort to Technology Service Corporation through the vehicle. In February 2026 the Joint Hypersonics Transition Office and NSWC Crane selected six performers for hypersonic development work: Leidos, Kratos SRE, Purdue Applied Research Institute, GoHypersonic, Halo Engines, and Aurex. A new hypersonic weapon system demonstration opportunity, sponsored by NSWC Crane with the Test Resource Management Center, was announced in January 2026.
Read that list carefully and two things stand out. Large primes take the integration roles on the biggest efforts. And the same lists carry university research institutes and small specialist engineering firms, which is where the nontraditional participation that opens the statutory gate actually comes from. The realistic entry for a smaller firm is a defined technical scope inside somebody else's integration, and that is not a consolation prize; it is how transition records get built.
Protest posture
GAO's bid protest jurisdiction comes from the Competition in Contracting Act and reaches procurement contracts. An other transaction is not one, so GAO generally declines to review the award. It will consider the narrower question of whether an agency improperly used other transaction authority in place of a procurement contract; MorphoTrust USA, LLC, B-412711 (May 2016), is the standard citation, and GAO denied that protest because the protester did not show the agency's use of the authority was inconsistent with its statute.
The Court of Federal Claims has moved. In Raytheon Co. v. United States, No. 24-1824C (Feb. 24, 2025), the court declined to dismiss a challenge to a Missile Defense Agency other transaction award, holding the decision was made "in connection with a procurement or proposed procurement." Judge Bonilla wrote that the court would be the "de facto forum" for bid protests involving these instruments. The practical read for 2026: the door at the Claims Court is open wider than practitioners assumed five years ago, it is still slow and expensive, and a consortium down-select is not a place to plan on litigating. Debriefs after a Request for Solutions are a courtesy, not a right.
How to read the vehicle before you spend the money
Four checks, in order, and none of them take a week.
Check the subject matter against the sponsor, not the manager. NSWC Crane buys microelectronics, spectrum, and strategic mission systems. If your capability is not one of those or a named component of one, the membership is cheap but the lane is wrong, and a different consortium or a different pathway will serve better.
Confirm your nontraditional status and write down the basis. The test at § 3014 is CAS coverage. Knowing which of the four conditions at § 4022(d)(1) your team opens is information the sponsor needs and most responses leave implicit.
Start long-lead registrations now. CAGE code and UEI for any vehicle. The DD-2345 if SpEC is anywhere in the plan. These are calendar items, not effort items, and they cannot be compressed when a request drops.
Find the program before the request appears. Requests are written around a requirement that already exists. The firms that respond well are the ones that met the technical office at an industry day, a consortium event, or through a research project months earlier. Membership gets the email. Nothing about membership gets the relationship.
Bottom line
NSTXL is a competent administrative layer that makes a Navy vehicle usable by companies that would never survive a full Part 15 competition, and it charges the government rather than the performer for doing it. S2MARTS is a real vehicle with a decade-long term, no stated ceiling, and a specific technical appetite. Together they lower the cost of entry to near zero and leave the hard part exactly where it was.
The hard part is a sponsor with an appropriation and a problem you can demonstrably solve. Consortium membership is the cheapest useful thing a firm can buy in this market and the least sufficient. Buy it, then go do the work that makes a request worth answering.
Frequently asked questions
No. It is a nonprofit consortium manager based in Arlington, Virginia. It holds base other transaction agreements with government sponsors, administers the membership and the resulting project agreements, and is paid a management fee by the government calculated as a percentage of project award value. It does not evaluate responses or select winners.
Yes. NSTXL states that membership is required to submit a response to any request. Membership activation is normally about one business day once the application is complete, but a CAGE code and Unique Entity ID have to exist first, and SpEC additionally requires a DD-2345 at registration.
Annual dues are tiered by entity type and revenue, starting at $250 for a corporate member under $10 million in revenue and rising to $10,000 for entities above $100 million. Universities and laboratories pay $2,500, incubators $500. One fee covers every NSTXL-managed vehicle, and participation limited to the S2MARTS Research lane carries no membership cost.
The published fast-end figures are about 70 days from request to award for S2MARTS and about 85 days for a Space Enterprise Consortium request for prototype proposals. Larger integrated efforts take considerably longer; the consortium's own case study for a Space Force ground command-and-control program describes under eight months. Treat the fast numbers as the floor for a tightly scoped project, not as a planning assumption.
Whatever the project agreement says. The DFARS data rights clauses do not attach automatically because an other transaction is not a procurement contract, and NSTXL states that IP and data rights are negotiated project by project before the final agreement, with commercial IP developed before the engagement remaining the vendor's property. The firms that come out well arrive with a dated background IP schedule already written.