SpaceX’s Raptor 3 engine, built for the upcoming Starship V3 vehicle, now produces 250 tonnes-force of thrust at sea level, according to figures the company has shared publicly. That marks roughly a 9 percent jump over the Raptor 2, which was rated at approximately 230 tonnes-force. The gain is not just about raw power: the Raptor 3 is a fundamentally redesigned engine, and its performance has direct implications for NASA’s Artemis III mission, the flight intended to return astronauts to the lunar surface for the first time since 1972.
What changed inside the engine
The Raptor 3 is not a simple tune-up of its predecessor. SpaceX stripped away the engine’s external heat shield and protective skirt, simplified the turbopump assembly, and reduced the overall part count. The result is a lighter, more compact engine that achieves higher chamber pressure while being easier to manufacture at scale. Elon Musk has described the Raptor 3 as approaching the theoretical performance limits of a full-flow staged-combustion methane engine, a claim that independent propulsion engineers have noted is ambitious but not implausible given the design’s architecture.
The math behind the 9 percent figure is straightforward. Raptor 2 engines delivered approximately 230 tonnes-force at sea level. At 250 tonnes-force, the Raptor 3 represents an increase of about 20 tonnes-force per engine. Multiply that across the 33 sea-level engines on a Super Heavy booster, and the aggregate thrust gain is substantial, potentially exceeding 650 additional tonnes-force at liftoff. For a vehicle already producing more thrust than any rocket in history, that margin could translate into heavier payloads, larger propellant reserves, or both.
Where NASA’s testing stands
NASA confirmed in a September 2023 missions blog post that SpaceX had completed a round of engine tests tied to the Artemis III Human Landing System (HLS). That campaign covered both sea-level and vacuum-optimized Raptor variants, the two configurations needed for different phases of a lunar mission. Sea-level engines power the Super Heavy booster off the pad; vacuum Raptors, with their larger bell nozzles, handle orbital maneuvers and the descent to the Moon’s surface.
Since that 2023 milestone, the broader Starship test program has accelerated. SpaceX conducted multiple integrated flight tests through 2024 and 2025, including the dramatic booster catch at the Starbase launch tower and progressively longer upper-stage flights. Each test has fed engineering data back into the Raptor development pipeline, though SpaceX has not publicly detailed which engine generation powered each mission.
NASA’s role in this process is that of a customer and safety authority. The agency reviews SpaceX’s test data under the HLS contract but has not published independent thrust verification figures, throttling profiles, or restart reliability statistics for the Raptor 3. That is not unusual for a program still in development; NASA typically reserves detailed performance disclosures for formal certification milestones or post-flight reports. But it does mean the 250 tonnes-force figure currently rests on SpaceX’s own public statements rather than independent government validation.
The tanker flight question
One of the most-watched variables in the Artemis architecture is the number of orbital refueling flights Starship will need before the HLS lander variant carries enough propellant to reach the Moon. The Starship V3 is designed to be physically larger than the V2, with a stretched propellant tank capacity reported at roughly 2,000 tonnes compared to the V2’s approximately 1,200 tonnes. That expanded capacity, combined with higher engine thrust, could meaningfully reduce the number of tanker launches required.
However, no NASA or SpaceX document has publicly quantified how many fewer tanker flights the V3 and Raptor 3 combination would require. The relationship between engine thrust, vehicle mass, propellant load, and mission trajectory is complex, and small changes in one variable can cascade through the entire flight plan. Claims circulating online that Raptor 3 “eliminates” a specific number of tanker flights should be treated as speculation until either organization publishes detailed mission planning data.
What is clear is that the refueling challenge extends well beyond engine thrust. Transferring cryogenic propellant between two vehicles in orbit has never been demonstrated at the scale Starship requires. SpaceX must prove that liquid oxygen and methane can be kept cold enough during transfer to avoid boil-off losses, and that the docking and fluid-coupling systems work reliably in microgravity. Higher thrust helps on the margins, but it does not solve the plumbing.
Timeline and what to watch
NASA’s current public target for the Artemis III crewed lunar landing has shifted several times and, as of early 2026, sits at no earlier than September 2026. That date is widely regarded within the aerospace community as optimistic, given the number of technical milestones still ahead: orbital refueling demonstrations, an uncrewed lunar landing test, and final crew-safety certification for the HLS vehicle.
There is also an open question about which Raptor generation will power the first crewed landing attempt. SpaceX has a history of mixing engine versions on early flights as newer designs mature, and NASA’s safety culture tends to favor well-characterized hardware over the latest revision. It is plausible that the first Artemis HLS mission could fly on Raptor 2 engines while Raptor 3 accumulates flight heritage on uncrewed Starlink deployment missions and cargo flights. No public certification plan has clarified this point.
Meanwhile, the competitive landscape is shifting. Blue Origin holds a separate NASA contract to develop a lunar lander for Artemis V, giving the agency a second provider and reducing its dependence on any single propulsion system. For SpaceX, demonstrating Raptor 3’s reliability is not just about Artemis; it underpins the company’s commercial business case for heavy Starlink satellite deployments and its long-stated ambition of sending cargo and eventually crew to Mars.
What the numbers do and don’t prove
The 250 tonnes-force figure is credible in context. It aligns with the trajectory of Raptor development, which has seen thrust climb from roughly 185 tonnes-force on early Raptor 1 units to 230 on the Raptor 2. The engineering changes SpaceX has described, particularly higher chamber pressure and reduced parasitic mass, are consistent with that level of improvement. But credible is not the same as independently confirmed.
For readers tracking Artemis III, the practical picture as of mid-2026 is this: SpaceX’s engine development is progressing through real test milestones, and NASA has publicly acknowledged that progress. The specific performance gains SpaceX attributes to the Raptor 3 have not yet been validated by independent data or flight results. Higher thrust is a meaningful advantage, but it does not by itself resolve the program’s harder open problems: orbital refueling at scale, cryogenic propellant management, and precision landing on the lunar surface.
Those answers will come not from press releases or social media posts, but from flight data. The next round of integrated Starship tests, expected in the coming months, should begin to close the gap between advertised performance and demonstrated capability. Until then, the Raptor 3’s 250 tonnes-force stands as a promising benchmark with real engineering substance behind it, awaiting the proof that only a full mission can provide.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
