On 28 September 2026, SpaceX's Starship lifted off from Starbase, Texas, reached orbit for the first time, and released 26 full-size Starlink internet satellites. That sentence contains three milestones at once: the largest rocket ever built finally completed the journey to orbit it was designed for, it carried a real payload rather than a test mass, and that payload was the next generation of Starlink satellites too large for Falcon 9 to launch. As of mid-October 2026, Starship had flown 14 test flights in total, and Flight 14 was the first to close the loop from launch to orbital deployment.

The vehicle that flew was Version 3, powered by 33 Raptor 3 engines on the Super Heavy booster and six Raptor engines on the ship itself — three standard and three vacuum-optimised for space. Stacked, the system stands about 124 metres tall, with a 72-metre booster holding roughly 3,650 tonnes of propellant and a 52-metre ship holding about 1,600 tonnes. Those numbers explain both the ambition and the difficulty: no rocket this large has ever been made reusable.

The sequence begins with all 33 booster engines igniting at Starbase and the vehicle climbing out over the Gulf of Mexico. After a few minutes, hot staging takes place: the ship lights its own engines while still attached to the booster, then pulls away — a technique that saves time and propellant compared with waiting for separation. The booster then flips, relights engines for a boostback burn to head home, and either lands in a tower catch or splashes down at sea, depending on the flight plan.

The ship, meanwhile, burns on to orbit, deploys its payload, and later fires its engines to come out of orbit for a controlled re-entry. On Flight 14 the deployment of 26 Starlink V3 satellites was the operational payoff: these larger satellites are central to SpaceX's plan for a higher-capacity Starlink network, and Starship is the only vehicle in its fleet that can carry them in quantity. A test flight that also earns its keep is a sign that a programme is moving from experiment to operations.

Starship's path to orbit was not a straight line. Earlier flights tested pieces of the puzzle separately: booster return and tower catches, hot staging, ship re-entry through the atmosphere, and heat-shield survival. Each flight that ended in a splashdown, a breakup or a lost stage still returned data that changed the next vehicle's design, materials or flight software. That iterative approach — fly, learn, change hardware, fly again — is why SpaceX could move from first flight to orbit in about three and a half years despite the vehicle's unprecedented size.

The cost of that iteration is real: reports put Starship development spending above 15 billion dollars, and SpaceX does not publish what a single test flight costs while stages are still being lost. The company's stated target — fully reused Starships cutting launch cost per kilogram by about 99 percent from the historical average — is a goal, not a current price. Flight 14 proves the vehicle can reach orbit with payload; it does not yet prove it can do so cheaply, repeatedly, or with both stages recovered every time.

Three capabilities remain unproven at operational scale, and they decide whether Starship becomes a workhorse or stays an impressive demonstrator. First, routine rapid reuse: both stages must be caught or landed, inspected quickly, and reflown without months of refurbishment — the aircraft-like operations model SpaceX aims for. Second, orbital refuelling: tanker Starships must transfer super-cold liquid methane and oxygen to another ship in orbit, a cryogenic transfer NASA's inspector general has flagged as one of the programme's biggest technical challenges.

Third, the lunar variant: a specialised Starship Human Landing System, roughly 50 metres tall, must carry astronauts between lunar orbit and the Moon's surface on Artemis missions. Estimates cited by NASA's Aerospace Safety Advisory Panel suggest around 22 Starship flights may be needed to support a single lunar landing mission, including tanker launches. Flight 14's orbital success makes those flights plausible; it does not make them routine.

Watch three indicators after Flight 14. First, flight cadence: does SpaceX fly again within weeks, or does the next flight slip by months? Second, recovery: are both stages recovered, and how quickly does a recovered booster fly again? Third, payload: do subsequent flights deploy more Starlink V3 satellites, and does the manifest expand to other customers or to Artemis test hardware?

If cadence rises, recovery becomes routine, and payloads grow, Starship will have crossed from test programme to launch system — and the economics of everything from satellite constellations to lunar missions will start to change. If flights remain sporadic and stages are still lost, Flight 14 will stand as a milestone rather than a turning point. Either way, 28 September 2026 is now the date the largest rocket ever built first did the job it was built for.