Tesla Roadster 2026: Can the New Flying Car Really Fly?

Tesla’s Flying Roadster Is Real. But Can It Actually Fly?

Tesla has shown a Roadster jumping, turning in midair and landing with help from high-pressure gas thrusters. The demonstration was real. Sustained flight, passenger safety and a production-ready car are different questions.

Tesla’s next-generation Roadster promises to push the boundaries of electric performance. Could its SpaceX-inspired thrusters bring us closer to flying cars? Concept illustration by Next Horizon.

A white sports car climbs a huge curved ramp at a SpaceX test facility in Texas. Near the top, it is suddenly driving upside down. Then it leaves the ramp, rotates through the air and lands on its wheels. In another stunt, it leaps over a line of Tesla vehicles. It looks like something from a superhero film. Tesla showed it happening on October 15, 2026.

The machine was the long-awaited second-generation Tesla Roadster. Its tricks were powered not by hidden wings or a miniature jet engine, but by two very different technologies: electric fans that pull the car toward the road, and gas thrusters that push it away. The demonstration itself was real, as documented by Reuters and Associated Press. What it did not show was a car cruising over a city like a small aircraft.

So has Elon Musk finally built a flying car? That depends on what we mean by flying. The Roadster can leave the ground and make controlled adjustments in the air. It has not demonstrated the ability to carry people from one destination to another through the sky. The interesting story is what Tesla and SpaceX engineers accomplished—and what the demonstration still cannot tell us.

A nine-year promise finally takes off

Tesla introduced a second-generation Roadster prototype in November 2017. The sales pitch was extraordinary speed, long range and an electric supercar intended to beat its gasoline-powered rivals. Deliveries were initially promised for 2020. The date passed, followed by years of revised expectations.

The delay matters because the company was not attempting a routine model update. In the meantime Tesla’s priorities shifted toward mass-market electric cars, energy storage, software, robotaxis and humanoid robots. If you want that bigger context, our earlier history and assessment of Tesla in 2026 explores how its business changed. The Roadster, meanwhile, became a promise waiting for a new reason to exist.

At McGregor, the company finally produced a dramatic new demonstration. The redesigned Roadster climbed an inverted ramp using suction, righted itself with gas jets and performed a separate leap through the air. Tesla also played footage of an acceleration test that it said showed 0–60 mph (0–96.6 km/h) in 0.81 seconds. These are company demonstrations under controlled conditions, not independent tests of a finished production car.

Tesla now targets production in 2028. After the project's long history of delays, that should be treated as a goal rather than a delivery guarantee.

First trick: how do you drive upside down?

Think of a suction cup stuck to a smooth window. It holds because the air pressure outside the cup is higher than the pressure inside. Now imagine keeping that pressure difference alive with electric fans instead of a one-time push. The Roadster uses four powerful fans and movable skirts underneath the car to remove air from the narrow space between its floor and the track.

Nothing here cancels gravity. Lower pressure under the vehicle lets the surrounding air push it toward the surface. On an ordinary road, that extra force helps the tires grip. On a specially shaped inverted track, it can help hold the car upside down—as long as the fans keep running and the skirts maintain a good seal.

Conventional race cars also use aerodynamic downforce, but their wings depend heavily on the vehicle moving through the air. Fan-generated downforce can work even at low speeds. The principle is proven: the British McMurtry Spéirling fan car used it to set a remarkable Goodwood hillclimb record in 2022. Tesla’s contribution is not inventing suction, but integrating an unusually powerful system into a proposed road-going hypercar.

Powerful underbody fans can create a low-pressure zone that helps hold a vehicle against the road — even when the road is upside down. This conceptual illustration explains the principle, not the exact Tesla engineering design.

Tesla engineering chief Lars Moravy described four 20-kilowatt fans and deployable underbody skirts in a technical walk-through reported by WIRED. The company says the system can support cornering forces exceeding 2 g. That figure describes the intensity of the sideways load during a turn; it is not evidence that a car can fly without a supporting surface.

Second trick: a rocket without fire

The airborne trick uses cold-gas thrusters. The idea is simple: let compressed gas rush out of a nozzle in one direction, and the vehicle is pushed the other way. A released party balloon darts around a room for the same basic reason. Tesla and SpaceX have turned that principle into a system of pressure vessels, valves and precisely aimed jets.

No fuel needs to burn in the nozzle. Stored pressure supplies the energy. When gas escapes at high speed, the vehicle receives an opposite push. This is Newton’s third law in practical form: every action has an equal and opposite reaction. NASA’s survey of small-spacecraft propulsion describes cold-gas systems as simple and mature technology, often using nitrogen for satellite attitude control. On a spacecraft, small jets can rotate the vehicle. On a car, much larger jets must fight Earth’s gravity.

The versions use this hardware differently. In Tesla's descriptions, Thrust adds an approximately 200-liter high-pressure vessel and acceleration jets behind the seats. The experimental Rocket configuration adds another 400-liter vessel and six additional jets, bringing the total to ten nozzles and about 600 liters of tank capacity. That larger tank takes the place of the passenger seat. During the reveal, Tesla engineers described roughly 250 kilograms of compressed gas in the full system.

A pressure of around 10,000 psi—about 690 bar—was cited for the smaller vessel. That is an extraordinary amount of stored energy, requiring carefully engineered tanks, valves and crash protection. Tesla kept people out of its aerial demonstrations. Former aerodynamic lead James Crook later explained that the high-pressure tanks were still experimental and carried significant safety risks.

Instead of burning rocket fuel, cold-gas thrusters release compressed gas to produce short bursts of thrust. This conceptual cutaway illustrates how such a system might fit inside a high-performance electric car.

Position those jets around the car and they can do more than push it forward or up. Different bursts can make its nose rise, roll the body or turn it sideways. In engineering terms, the jets produce both force and torque—the turning force that lets a machine change orientation. Computer control can use those pulses to help the car point its wheels down before landing.

The five-second problem

But a brief push is not the same as sustained lift. A cold-gas system works by throwing away the gas it carries. Every dramatic burst spends some of a finite supply. Unlike an electric rotor that can keep moving surrounding air, the Roadster must take its pressurized working gas along for the ride and refill its vessels after use.

That constraint is severe. Reporting based on Tesla engineering briefings indicates that maximum thrust is available for only a few seconds. In the prototype described to WIRED, refilling the complete 600-liter flight system takes roughly six to seven hours. Engineers hope to cut that to around 30 minutes. Neither number describes an established customer-ready refilling network.

Then there is weight. The car needs a battery, motors and all the normal equipment of a road vehicle. Tanks, compressed gas, pipes and valves pile on more mass—exactly what a flying machine wants to avoid. More mass requires more thrust; more thrust usually calls for still more gas and hardware. It is an engineering loop that quickly becomes expensive and difficult.

Consider what a commuter would actually need. A short hop over an obstacle is spectacular, but it is not a practical way to beat traffic. Routine airborne travel means lifting passengers reliably, coping with wind, controlling the landing, maintaining safety after a malfunction and repeating the trip. No public Roadster test has demonstrated that package.

How fast is the new Roadster?

The Roadster makes more sense, at least for now, as an extreme electric sports car with a remarkable stunt mode. Tesla says the Performance model will target 0–60 mph in about 1.6 seconds and around 210 mph (338 km/h) at the top end. Thruster-equipped versions are promised even faster launches. In separate Tesla-shown footage, a development vehicle reached 60 mph in a claimed 0.81 seconds. The company also discussed a top speed exceeding 230 mph (370 km/h) and a range of more than 300 miles (480 km). The figures refer to different configurations and remain manufacturer claims.

For perspective, reaching 60 mph in 0.81 seconds means an average acceleration of roughly 3.4 g—more than three times the acceleration due to Earth's gravity. That is extraordinary. But it is still a figure from Tesla's own test footage, not an independently certified production-car record. Timing methods, tires, surfaces and the exact configuration all affect comparisons.

Musk also spoke about reaching half the speed of sound: roughly 380 mph (610 km/h), depending on air temperature. That is far beyond the quoted 230-mph target. It is a wish, not a demonstrated specification. With a car this unusual, the distinction between what has been filmed, what has been measured independently and what is merely planned is essential.

Three versions, three different promises

Tesla’s reservation page lists the Roadster in Performance, Thrust and Rocket forms. Performance is presented as the road-focused electric sports car with fan-based downforce. Thrust adds acceleration jets and is positioned around track performance. Rocket is described as a single-seat, invitation-only model with a full flight system, and Tesla states it is not approved for use on public roads.

Performance currently requires a $50,000 reservation deposit; Thrust requires $250,000. WIRED reports that each starts with $5,000, with the balance due within ten days. Those amounts reserve a place in line—they are not published purchase prices. Rocket is invitation-only, and Tesla has not provided a guaranteed customer delivery date.

A reservation page can establish that Tesla is accepting commitments. It does not tell us how close its most complex prototype is to safe, repeatable production. The distinction matters especially for the one-seat Rocket model with its additional tank and flight hardware.

A controlled jump is an impressive engineering achievement, but it is not the same as sustained flight. The challenge is turning short bursts of thrust into safe, practical transportation. Concept illustration by Next Horizon.

Is this actually a flying car?

The simplest test is this: could the vehicle transport someone through the air between two addresses? Tesla has shown a Roadster jump, rotate and land. It has not shown sustained, steerable travel carrying a passenger, nor routine operation in wind and bad weather. A few seconds in the air and a useful aircraft are very different engineering problems.

There is also a legal wrinkle. In reporting by WIRED on the regulatory questions, an FAA spokesperson said that the demonstrated Tesla vehicle does not meet the agency’s definition of an aircraft. That is a statement about this specific vehicle, not general permission to jump a car above city streets. Road-safety standards, local traffic rules, track rules and the regulation of future aerial operations still matter. Musk himself acknowledged uncertainty about whether the Roadster would be lawful in every jurisdiction.

This is where the Roadster differs from the companies building electric vertical-takeoff aircraft. An eVTOL air taxi uses wings or multiple propellers to generate sustained lift and is designed from the beginning as an aircraft. A Roadster remains a road car with specialized systems for short bursts of motion. Our guide to flying cars and eVTOL air taxis explains why that distinction is fundamental.

Dismiss the demonstration as fake, and you overlook a real piece of engineering. Treat it as the arrival of everyday flying cars, and you overlook the limits. Both reactions make the story less interesting than it really is.

The engineering story behind the spectacle

One of the revealing accounts appeared on October 19. In The Times, former Tesla aerodynamic lead James Crook described the work as Project Mede—short for “Most Epic Demo Ever.” Engineers had to solve conflicting problems: strong suction to keep the car attached to a surface, low weight and drag to make it quick, and powerful gas jets to launch and stabilize it in the air.

Crook's account also shifts attention from the person on stage to the team behind the car. Aerodynamicists worked alongside pressure-vessel specialists, vehicle designers, control engineers and SpaceX collaborators. They built prototypes, made mistakes and had to balance demands that normally belong to very different machines.

Cost, Crook said, was not the project's main constraint. The priority was making the demonstration possible. That freedom can produce unusual engineering—but it does not answer what a customer would pay, how often the jets can be used or how the system behaves over years of ordinary ownership.

What happens next—and what might actually matter

The next meaningful milestone is not another dramatic video. It is repeatable testing with publicly explained conditions: passenger-carrying safety, a practical refill process, independent performance measurements and a clear account of which versions could legally be used on roads. A credible path to manufacturing matters too. Until Tesla supplies that evidence, 2028 remains a target.

There may still be lasting value in the experiment even if nobody ever commutes in a rocket-assisted Roadster. Lighter pressure vessels, faster valves and more precise vehicle controls could prove useful elsewhere. Fan-assisted downforce already has applications in racing. That does not mean every Roadster component will migrate into everyday cars; it means the engineering is worth separating from the sales pitch.

The timing also tells us something about Tesla. As the company invests in vehicles that might drive themselves, it has revealed perhaps its most extreme vehicle for a human driver. Musk has described the Roadster as the last of Tesla’s great manually driven cars. That makes it both a celebration of driving and an oddly fitting farewell to the era when carmakers primarily competed to make the driver faster.

I find the combination more revealing than the stunt itself: a system borrowed from spacecraft helps a sports car leap, while a racing idea holds it upside down. Both work, for a moment, under carefully prepared conditions. The next chapter will be written by the less glamorous tests—repeatability, maintenance, safety and whether any of it can become useful outside a test track.

Beyond the spectacular demonstrations lies the real challenge: making extreme performance repeatable, reliable and safe. Tesla targets Roadster production in 2028, but significant engineering questions remain. Concept illustration by Next Horizon.

The verdict: real physics, unfinished future

The new Roadster can stick to a purpose-built inverted ramp and complete short, controlled airborne manoeuvres. It is a striking demonstration of familiar physics combined in an unfamiliar way, and the engineers deserve credit for getting it to work.

What it is not—yet—is proof of a safe flying car for customers. We have not seen sustained travel, human passengers in the aerial test, routine refilling or a confirmed production model that can perform every demonstrated trick. For now, Tesla has given us an extraordinary experimental vehicle, not an alternative to the road. That is still a story worth telling—provided we tell it precisely.

Frequently asked questions

Did Tesla really make a car fly?

Tesla demonstrated a prototype leaving the ground and using gas jets to control short aerial movements in October 2026. That is not the same as sustained travel between destinations, and no human was inside during the aerial stunts.

Does the Roadster use rocket fuel?

No conventional rocket fuel burns in its cold-gas thrusters. The system releases highly compressed gas through nozzles. Engineers have discussed nitrogen in technical interviews, while the reveal also described its supply as compressed air.

Can the Roadster drive upside down?

Tesla showed its prototype moving upside down on a specially built ramp, using fans to maintain a low-pressure area underneath. It has not shown that the car could attach safely to an arbitrary ceiling or public structure.

When will the Tesla Roadster be available, and how much will it cost?

Tesla is targeting 2028 production, but has not confirmed a retail price or guaranteed deliveries. Deposits currently listed for reservations are not the price of the car.