Concorde vs Hyperloop: how the retired supersonic jet is still faster

Elon Musk's 2013 Hyperloop proposal promised a pod-in-a-tube faster than any airliner. Checked against the record books, it still is not faster than Concorde — a jet that has not flown a single passenger since 2003.

Concorde and Hyperloop belong to different categories of fact. Concorde is history: a supersonic airliner that entered service in 1976, carried fare-paying passengers across the Atlantic for 27 years, and was retired by Air France and British Airways in 2003. Hyperloop is a proposal: a white paper Musk published in August 2013 sketching a pod riding on air bearings through a near-vacuum tube, with no working full-scale version built since. Comparing their top speeds means comparing a flown number against a design target.

Top speed, checked against the record books

Concorde's cruise and maximum speed was Mach 2.04, about 2,180 km/h (1,354 mph). That ceiling was not arbitrary: at sustained supersonic speed, air compression heated Concorde's aluminium skin, and if the nose passed roughly 127°C the crew had to throttle back. Going faster than Mach 2.04 for a commercial lifetime would have meant building in titanium instead of aluminium, at a cost the project could not carry.

Hyperloop's alpha-paper proposal, by contrast, targeted a pod top speed of 760 mph (1,220 km/h) for aerodynamic efficiency inside the tube, on an illustrative Los Angeles-to-San Francisco route of roughly 560 km. It has never been demonstrated at that speed, or at passenger scale, by any company.

• Concorde cruise/top speed: Mach 2.04 — about 2,180 km/h / 1,354 mph (flown, retired 2003)
• Hyperloop proposed top speed: 1,220 km/h / 760 mph (2013 concept, never built)
• Boeing 777-200LR cruise speed: about 905 km/h (flown, currently in service)
• Shanghai Maglev commercial top speed: 431 km/h (operated 2004–2021, since reduced to 300 km/h)

The circumnavigation question: what the record books actually say

Concorde's most-cited round-the-world flight departed New York’s JFK on 15 August 1995, refuelling at Toulouse, Dubai, Bangkok, Guam, Honolulu and Acapulco before returning to JFK. The official elapsed time was 31 hours, 27 minutes, 49 seconds, covering roughly 40,600 km (25,252 statute miles) — a record that still stands as the faster of Concorde's two circumnavigations. A separate, slower record of 32 hours 49 minutes 3 seconds belongs to an earlier, 1992 flight that departed Lisbon on a different routing; the two figures are not the same flight.

Of that 31.5 hours, only 22 hours 39 minutes 21 seconds was airborne, at an average speed of 1,114 mph (about 1,793 km/h) — lower than Concorde's Mach 2.04 cruise because supersonic flight was banned over land and much of the route was flown subsonic. The rest, nearly nine hours, was spent on the ground refuelling at six stops.

Even the most generous hypothetical for Hyperloop — no stops, constant top speed, the entire way around the globe — still finishes behind Concorde's actual 1995 record, refuelling stops and all.

Applying Hyperloop's proposed 1,220 km/h top speed to that same 40,600 km distance, with no stops and no time lost accelerating or decelerating, gives a notional time of about 33 hours 20 minutes — nearly two hours slower than Concorde's actual, stop-included record of 31 hours 28 minutes. That gap would only widen once realistic station stops, curves and ramp-up time were factored in for a system that has never carried a passenger.

Why Hyperloop never got a chance to prove it wrong

The comparison above is not entirely fair to Hyperloop, and it is worth saying so. Musk's 2013 paper never proposed circling the globe — its worked example was a single 560 km commuter corridor, not intercontinental travel, and building a route of any length has proved far harder than the paper assumed. Early hyperloop cost estimates in the hundreds of millions of dollars gave way, once real feasibility studies were done, to estimates of $20–40 million or more per kilometre of track — several times the original projections.

No company ever closed that gap. Hyperloop One, the best-funded effort, raised more than $450 million but never signed a contract to build a working system before shutting down in December 2023. By 2026, competitors fared no better: the Netherlands’ Hardt Hyperloop was declared bankrupt in March 2026, and Spain’s Zeleros entered insolvency, its assets sold off. As of 2026, no operational hyperloop system exists anywhere in the world.

What actually ended Concorde

Concorde's retirement in 2003 had nothing to do with being overtaken on speed. High fuel burn, a supersonic-fare market that shrank sharply after 2001, and rising maintenance costs on an ageing fleet made the aircraft unprofitable to keep flying, and Air France and British Airways withdrew it that year after 27 years of commercial service. Nothing built since has matched its cruising speed for a paying passenger.

Frequently asked questions