National Security & DefenseAI & Data ScienceMarkets & Finance

U.S. Space Force Adds New Startups as It Cuts Reliance on SpaceX

Two startups join the $5.6 billion launch pool, one without a rocket.

Josh LynwoodFounder
Published
Read time
11 min
Share
Where Things Stand

The $5.6 Billion Is a Competition License, Not Revenue Awarded

The Space Force added two companies to a launch pool, a decision that did three things at once. It broadened the national-security launch base for the third year running, using a routine annual The mechanism that reopens the launch contract each year to add newly qualified providers to the pool. as an industrial-base instrument. It admitted a company that owns no rocket, only the stage that maneuvers a payload to its final orbit, as a launch prime for the first time in the program's history. And it handed two venture-backed companies, one founded by SpaceX's first employee and propulsion lead, Tom Mueller, and one now run by former Google chief Eric Schmidt, a national-security launch position. The through-line, that a deliberate concentration bet, a redefinition of launch, and a capital event are one event, is an analytical judgment, not a claim the Space Force has made.

  • The Award. On July 7, 2026, per the Department of War contracts notice, Impulse Space (Redondo Beach, California; contract FA8811-26-D-B001) and Relativity Federal (Long Beach, California; FA8811-26-D-B002) were awarded multiple-award, firm-fixed-price, indefinite-delivery/indefinite-quantity (An indefinite-delivery, indefinite-quantity contract, a vehicle that sets a shared ceiling and lets the government order work later through task orders.) contracts with a maximum cumulative ceiling of $5.6 billion for National Security Space Launch, the Space Force program that buys launches for military and intelligence satellites. Phase 3 Lane 1. The acquisition was competitive with three offers received, and $10 million in fiscal 2026 space-procurement funds was obligated at award. Space Systems Command, Space Access, Los Angeles Air Force Base is the contracting activity.
  • The Money, Precisely. State the money as $10 million total ($5 million each) against a $5.6 billion ceiling. Per the Space Systems Command (SSC) release, Impulse and Relativity will each receive a $5 million firm-fixed-price task order to conduct an initial capabilities assessment and develop their approach to tailored mission assurance. That is a study order, not a launch order, so the $5.6 billion is a competition license the pool shares, not revenue awarded to these two firms.
  • The On-Ramp Arc. This is the third annual on-ramp in a row, and the pool went from three to five to seven. SSC and CRS confirm the sequence: Blue Origin, SpaceX and ULA were the inaugural Lane 1 providers (FY24); Rocket Lab and Stoke Space were added (FY25, March 2025); Impulse and Relativity were added July 7, 2026 (FY26). A deliberate, repeatable base-broadening, not a one-off buy.
  • The Statute Behind It. The assured-access mandate is 10 U.S.C. 2273, which sets U.S. policy to maintain at least two launch vehicles able to deliver any national-security payload. Congress pushed for more: the Senate Appropriations report S.Rept. 118-204 encouraged the Space Force to include a greater diversity of providers and more competition in Phase 3. SSC calls NSSL the single largest procurement activity in the U.S. Space Force and manages a $15.6 billion annual space-acquisition budget.
  • The Capability Isn't Flying Yet. The vehicle Impulse brings to NSSL is Helios, a high-energy orbital-transfer stage that is largely unflown, with first flight targeted 2027 (slipped from an earlier 2026 goal); its Deneb engine (LOX/methane, oxygen-rich staged combustion) has not flown. Impulse's real flight heritage is the smaller Mira tug. Relativity's NSSL vehicle, the Terran R rocket, is also pre-first-flight. Lane 1's qualification bar, a completed launch or a credible plan to do one, is doing real work here.
Sources5See all 20
Is the Space Force actually buying resilience, or just a bigger bench?

Seven Providers, and Assured Access Still Rests on Three

The pool moved. Lane 1 went from three providers to five to seven across three consecutive annual on-ramps: Blue Origin, SpaceX and ULA at the start, Rocket Lab and Stoke Space in March 2025, Impulse Space and Relativity Federal on July 7, 2026. The statute behind it, 10 U.S.C. 2273, sets US policy to maintain at least two launch vehicles able to deliver any national-security payload, and the Senate Appropriations report encouraged greater diversity of providers. But assured access for the hardest missions runs through Lane 2 certification, and that base is still two to three providers. Nothing in a Lane 1 on-ramp moves it.

Neither new entrant has flown the vehicle it was admitted on. Impulse's Helios, a high-energy orbital-transfer stage, is largely unflown with first flight targeted for 2027, slipped from an earlier 2026 goal, and its Deneb engine has not flown; the company's real flight heritage is the smaller Mira tug. Relativity's Terran R is also pre-first-flight. Lane 1's qualification bar, a completed launch or a credible plan to do one, is doing real work here. The money says the same thing. Each firm received a $5 million firm-fixed-price task order for an initial capabilities assessment, $10 million obligated against a $5.6 billion shared ceiling. A ceiling is a competition license, not revenue.

None of that makes the on-ramp cosmetic. Admitting a company that builds no rocket, only the stage that maneuvers a payload to its final orbit, changes what the program buys: trajectory and destination, not just ascent. That is a real redefinition, and it arrived before the civil regime authorizing such maneuvers is settled. But a bench becomes resilience only when someone comes off it. What would show that wrong: an entrant on-ramped in 2025 or 2026 wins a Lane 1 task order in competition against SpaceX or ULA, or a Lane 1 provider is invited toward Lane 2 certification. Either would move the base rather than the roster.

Sources12See all 20
Intersections

The government broadened the bench, deliberately

National Security & Defense. This is the third annual on-ramp under a statute (10 U.S.C. 2273) that requires at least two launch vehicles for assured access; the Space Force is engineering redundancy so no single provider is a single point of failure. AEI's Todd Harrison names the driver, a wariness about over-dependence on SpaceX and concern that SpaceX seems to be pivoting away from its traditional launch business, an attributed analyst read, not an official finding.

Regulatory Policy. The instrument is acquisition design, not a new program: Lane 1 is a multiple-award IDIQ under FAR Part 16.505, reopened annually to on-ramp new providers and off-ramp non-compliant ones, executing the Senate Appropriations directive for greater diversity of providers.

What would make this wrong

No provider is on-ramped in the FY27 cycle and the annual reopening lapses. An industrial-base instrument executing a statutory mandate does not quietly stop after three years of it.

Open question

On the acquisition side: the lane reopens every year and can off-ramp as well as add, so what has ever actually caused a provider to be removed from the pool?

Sources3See all 20
The Weave

The Weave maps a single development across domains and across time. Each row follows one domain from where things stand now through the next eighteen months, and expands for the reasoning behind that trajectory.

Wiiver
SECTOR / DOMAINclick a domain to expand
As It Standsthe current status
Immediate0–6 months
Near-Term6–18 months
Government + Policy
The Lane 1 pool stands at seven providers after a third straight on-ramp. Lane 2, the hardest missions, is untouched.
Launch prime, redefined
For the first time, NSSL admitted a launch prime that builds no rocket, an orbital-transfer vehicle (Impulse's Helios), a definitional shift, not a routine vendor add.
Deliberate SpaceX de-risking
This is the third annual on-ramp in a row (three to five to seven), engineered redundancy in assured access under a two-vehicle statute.
Technology + Engineering
Impulse's autonomy record is one camera-only rendezvous demo, closing to about 1,250 meters with a partner in control.
Autonomy enters the prime tier
By making Impulse a prime, the Space Force admitted an autonomous guidance and RPO stack, not just a rocket, into the national-security tier.
The GEO RPO demo is the near read
The near-term test is a late-2026 GEO rendezvous demo pairing an Anduril infrared imager and Lattice AI processor with Impulse's Mira, not the 2027 Helios flight.
Business + Markets
The prime slot sits five weeks after a $500 million Series D at a reported $4.26 billion post-money valuation.
Prime slot, weeks after a raise
At $5 million obligated, the prime slot is a credibility event, not a revenue event, landing five weeks after Impulse's $500 million Series D.
The prime slot de-risks the next raise
The prime slot is a capital-access and de-risking event, not revenue: it validates a mark struck weeks earlier and hands the next raise a national-security moat.
wiiver.co · 3 impacted domains shownWiiverv1 · July 10, 2026
Looking Forward

First Flights, Not the Pool Count, Decide If This Was Real

Whether this on-ramp becomes a real posture change or stays a paper precedent will show up in a handful of checkable signals over the next several quarters.

  • First Flights, not the Pool Count. Whether Helios flies in 2027 and Terran R debuts on schedule (targeted late 2026); a further slip, or a Deneb development stumble, converts the first-upper-stage-prime precedent into a paper capability, because a task order that actually flies a national-security payload is the signal, not pool admission.
  • The First Contested Task Order. Whether an on-ramped 2025 or 2026 entrant ever wins a Lane 1 task order in competition against SpaceX or ULA, likely no earlier than FY27 and gated on first flight; until then the $5.6 billion ceiling is an option, not a backlog, and the on-ramp is paper diversification.
  • Lane 2, not Just Lane 1. Whether any Lane 1 provider is ever invited toward Lane 2 certification; assured access for the hardest missions still rests on a two-to-three-provider certified base, and the broadening at Lane 1 does nothing for that concentration until a new entrant certifies up.
  • The Rules Catching up to the Capability. The FCC Space Modernization vote on July 22, 2026 (an ephemeris-sharing mandate, the nearest dated legal marker), and whether the Office of Space Commerce's voluntary on-orbit-servicing certification gets a statutory spine; procurement has admitted a maneuver-capable prime before the civil regime that authorizes such maneuvers is final.
  • The Tell. If first flights slip and the annual on-ramp keeps enlarging a pool SpaceX keeps winning, this was optics and the real provider base stays about three. If new entrants fly and start winning task orders while the on-ramp reallocates missions, the industrial-base bet worked and the Space Force really did buy resilience.
Sources6See all 20

Every issue

  1. 01Intersection of the week
  2. 02Impact of the week
  3. 03The week in review

Wiiver Weekly

One free email, Saturdays at 7:00 AM ET.

Unsubscribe anytime.

Sources and Verification
10 of the 20 sources cited here are primaryfilings, opinions, statutes and agency releases read directly
Primary sources10
Secondary sources, by sector10
Government + Policy2
Business + Markets4
Technology + Engineering4
v2 · Reviewed by Josh Lynwood · July 10, 2026
Back to top