Third-quarter 2026 earnings call, November 10, 2026

Firefly Aerospace used its third-quarter call to detail a three-year intellectual property, supply and services agreement with AST SpaceMobile valued at more than $1.3 billion. It also reported revenue of $110.7 million, up 259% year over year, and disclosed that Alpha Flight 8 will now fly in the first quarter of 2027 rather than before year-end. The company raised the low end of its 2026 revenue guidance to a range of $440 million to $450 million from $420 million to $450 million, lifting the midpoint to $445 million from $435 million.

The AST SpaceMobile Deal: Licensing Miranda, Not Selling Eclipse

CEO Jason Kim described the agreement as a non-dilutive collaboration that gives Firefly funds to expand launch production, test capacity and its supply chain. Firefly will build, test and deliver 27 Miranda first-stage engines and 9 Vira upper-stage engines, making up three complete flight sets. They will power AST's new reusable, three-stage heavy-lift vehicle, designed to carry 20 metric tons to low Earth orbit. Kim stressed that Firefly keeps full ownership of the engine intellectual property and is licensing it to AST, which will configure the engines for its own vehicle. "When a company sets out to build its own rocket, the engine is the hardest, highest stakes decision they make. AST chose Firefly, and that tells you everything you need to know about where our technology stands," Kim said.

CFO Darren Ma split the $1.3 billion into three parts. About $800 million is associated with the Miranda and Vira IP licenses, comprising $400 million in cash and 10 million to 14 million shares of AST Class A common stock, with the share count set by a forward VWAP calculation and subject to lockup periods. A further $250 million is for support services, and the remaining $250 million is for delivery of engines for AST's first three flights. Ma said everything other than the equity will flow through operating cash.

On revenue recognition, the IP and services portions will be recognized over time on a percentage-of-completion basis, and engine deliveries will be recognized at a point in time on delivery. All of it will sit in the launch segment. Ma said the program starts immediately but revenue begins in earnest in 2027, with engine deliveries ramping in year two and revenue rising through year three. In response to a question from Deutsche Bank about the typical 3- to 5-year slippage in new rocket programs, Ma said a delay on AST's side should not affect Firefly's revenue recognition, since the IP is tied to services performed over the three-year period and engines are recognized on delivery. Ma said gross margin is currently in the mid-20s and that the IP and services revenue will expand it, though he did not quantify the uplift.

The offsetting cost is investment. Ma said Firefly expects significant R&D, capital expenditure and infrastructure spending tied to the agreement, and that cash burn should "tick up through next year." Kim said the funds will pay for multiple new test bays at the Rocket Ranch, bulk purchases of long-lead materials, additional launch program staff and more integration and test capacity, all of which he said will also support Eclipse.

Analysts probed what Firefly gives up. Barclays asked whether AST could eventually build engines itself at scale without further revenue to Firefly, given the upfront $400 million license payment. Kim did not address royalties or production economics directly, saying AST "is going to eventually build the engines themselves" for its own launch vehicle and that the deal "does not change our identity or our business model." On restrictions, Kim said AST is focused solely on launching its own constellation while Firefly serves commercial and government customers with Eclipse. Asked by B. Riley why AST's design uses three stages when Eclipse uses two, Kim said that is AST's design and that AST is focused on constellation mass per launch, while Eclipse is built for a broader set of missions, including lunar ones.

Alpha Flight 8 Moves to First Quarter 2027

Kim said direction from a national security customer required Firefly to readjust its launch manifest, pushing Alpha Flight 8 into the first quarter of 2027. "We did not make this decision lightly as our preference was to launch before the end of the year," he said, adding that Firefly made a similar manifest change before the VICTUS NOX responsive launch on Flight 3. Flight 8 is the debut of the Block II configuration, which is designed to increase manufacturing rate and reliability. The company will review post-flight data before subsequent launches, and Kim said timing remains subject to customer readiness, range availability and weather.

Firefly declined to give a 2027 launch target, saying it will provide one on its fourth-quarter call. Kim described the factory as ramping. Structures and tanks for Flight 9 are complete, with all four Reaver engines and the Lightning engine in assembly. Flight 10 structures have completed proof testing, Flight 11 tanks and structures are finished, and Flight 12 composites are in production. The company has 24 Reaver engine chambers in production, which Kim put at about six shipsets. He said two Reavers, one Lightning and one Miranda qualification engine were hot-fired within three hours on the day of the call.

Kim said Alpha demand is the most supply-constrained in company history, with more customers than rockets in the near-term manifest. New awards in the quarter include two launches from Esrange in Sweden for SSC Space, Firefly's first European launch contracts, and up to 10 launches through 2030 for Japanese SAR operator Synspective at two satellites per mission. Firefly also added a third hypersonic task order from a confidential customer and extended its Lockheed Martin multi-launch agreement. On pricing, Kim pointed to the $22 million VICTUS SOL contract and said the company is "seeing things with upside opportunities beyond" those prices. He said demand supports higher pricing for medium lift as well.

Eclipse and Miranda Qualification

Miranda has entered its qualification campaign. Firefly has completed more than 175 hot-fire tests totaling 4,230 seconds of cumulative run time across 20 engines, including testing of multiple re-lights and power throttling for reusability. Kim said the company can now build and assemble Miranda engines for the first Eclipse launch, which it is developing with Northrop Grumman. He said he is adding technical experts to the program team and investing in infrastructure to ramp production. Asked about a first-launch date in 2027 and the remaining long poles, Kim did not confirm a date, saying the team is focused on qualifying Miranda, the composite tanks and the fluid systems and on delivering the first stage to Northrop.

On National Security Space Launch certification, Kim said Firefly wants to be able to tell the customer it is launching in a given year before it onboards to the program. He said the first stage is common to both Firefly's and Northrop's vehicles, and that Northrop's co-development gives confidence the rocket will meet national security and NASA standards.

Financial Results and Liquidity

Third-quarter revenue of $110.7 million compared with $30.8 million a year earlier, and year-to-date revenue reached $309.2 million, up 203%. Spacecraft contributed $106 million and launch $4.7 million. Ma said the quarter had no benefit from the accelerated national security hardware order that lifted the second quarter. GAAP gross margin was 22.9% versus 20.3% in the second quarter, with the improvement coming from mix after lower-margin hardware pass-through purchases weighed on the prior quarter.

Operating expenses rose to $104 million from $94.3 million, reflecting R&D on Alpha Block II and Eclipse. Operating loss was $78.7 million versus $70.4 million, net loss was $75.4 million, or $0.45 per share, and adjusted EBITDA was a loss of $69.5 million versus a loss of $61.2 million. Stock-based compensation was $15.3 million. Free cash flow was an outflow of $85.3 million, compared with $106.3 million in the second quarter, which included roughly $24 million for the final SciTec acquisition payment. Capital spending fell to $15.6 million from $24.7 million as Block II test upgrades and the spacecraft clean room expansion were completed.

Total liquidity was $839.7 million, made up of $534.7 million in cash, cash equivalents and short-term investments and $305 million of undrawn revolver capacity. Backlog rose to a record $1.7 billion from $1.5 billion in the prior quarter and $1.3 billion a year ago, helped by the ground-based radar digitization award, additional hardware orders and another Alpha hypersonics task order. Ma said more than $300 million of new contracts replenished the revenue recognized in the quarter. The ending share count was 171.6 million, which Ma expects to rise by about 1 million per quarter.

Asked to bridge the guidance change given Flight 8's slip, Ma said strength in spacecraft, five moon programs running in parallel and strong demand in AI software are offsetting the launch revenue impact. He did not say whether any Alpha launch is assumed in the fourth quarter. On launch gross margin, Ma said Alpha costs are expensed to R&D until the vehicle reaches technical feasibility, after which they will be inventoried, and Eclipse will follow a similar pattern.

Lunar Business: CLPS 2 and Five Missions in Parallel

Firefly has onboarded to NASA's $10 billion CLPS 2 contract, which Kim said lets it compete for task orders for larger cargo landers able to carry items such as light terrain vehicles. He said Firefly began design work on the larger landers ahead of time. Kim suggested Space Force could account for 10% of lunar-bound payloads, potentially rising to 25%, and said the Space Force has issued requests for information covering up to five satellites to survey cislunar space. He framed the opportunity as larger than the $30 billion NASA Moon Base program.

Firefly is executing five moon missions in parallel. Blue Ghost Mission 2, Rider 2 The Dark, completed proof testing and has its lander and Elytra transfer vehicle stacked and powered on, in a stack 22 feet tall carrying payloads from the U.S., Australia, Canada, the UAE and the UK. The company expects it to launch in 2027, depending on delivery to Florida and a confirmed launch window. The Gruithuisen Domes mission is in flight hardware production with launch targeted for 2028, and the Lunar South Pole mission completed its critical design review. New commercial customers include Starcloud, which will fly an AI compute payload on Elytra using data from Firefly's Solux autonomous vision system acquired with Space-ng, and Zeno Power Systems, which will fly a radioisotope heater payload on the Near Side mission.

Kim said at least three CLPS 1.0 opportunities remain outstanding. One, a far-side south pole mission, could be awarded as early as the fourth quarter. A second is a commercial lunar surface imaging service similar to Firefly's Ocula offering, and a third is a landing mission likely to be released in early 2027 and awarded later that year. The newly opened clean room in Austin quadruples assembly and test space and is designed to integrate up to a dozen landers and orbiters at once.

Elytra and SciTec: Orbital Services and Defense Software

On Elytra, the MoonFall mission, a $75 million subcontract to deliver NASA JPL drones to low lunar orbit, passed its preliminary design review. Firefly added a Defense Innovation Unit contract for an orbital services mission, won a spacecraft subcontract on NASA's Skyfall Mars mission, and is moving Project Sinequone into integration and testing ahead of a launch next year. Comparing Elytra with True Anomaly, Quantum Space and Turion, Kim cited large chemical propulsion fuel capacity, carbon composite structures that free up payload mass, and Space-ng's autonomous navigation cameras for rendezvous and proximity operations.

At SciTec, Kim described the FORGE missile warning and tracking program as an $800 million program "and counting" that is operational and performing. He said Firefly won the roughly $94 million first phase of the Ground-Based Radar Digitization program by lifting and shifting the FORGE framework, and that the program carries potential upside of about $400 million to $500 million. Firefly also onboarded to the $981 million NITE-STAR IDIQ and a $66 million instrumentation and testing IDIQ, and extended its Air Battle Management System work, which received Air Force interoperability test certification. Kim said the shift toward firm-fixed-price OTA contracts offers gross margin expansion if Firefly executes. On M&A, he said the company has "a robust pipeline of M&A targets" as part of a multiyear strategy.