Mars plans
The reason SpaceX exists. The 2026 launch window is open right now — and for the first time since 2016 the company has said, out loud, that the Moon comes first. Here is exactly where the Mars plan stands, how it got here, and what still has to happen.
SpaceX was founded in 2002 because Musk wanted a greenhouse on Mars and nobody would sell him a cheap rocket to put it there. Twenty-four years later the rocket exists and has reached orbit. The Mars part is still the hardest thing anyone has tried.
Where things stand, October 2026
No Starship is flying to Mars this window. On February 6, 2026 the Wall Street Journal reported SpaceX had told investors it would “prioritize going to the moon first” and attempt Mars later, with an uncrewed lunar landing targeted for March 2027. Two days later Musk confirmed it on X: “the overriding priority is securing the future of civilization and the Moon is faster.” A Mars city effort, he said, would begin in roughly five to seven years. WSJ via Yahoo KSL/Reuters
Mars was not mentioned on either 2026 earnings call or in the Q3 results. Space.com's October 14 assessment, marking Starship's tenth birthday: a 2026 attempt is “off the table.” Space.com
Why not?
Two reasons, both technical. First, Starship only reached orbit on September 28, 2026 — too late to prepare and refuel a Mars-bound fleet before the window closed. Second, orbital refueling has never been demonstrated between two ships, and a Mars ship needs about 1,200 tonnes of propellant loaded in orbit. Shotwell targets the first refueling test for early 2027.
How the date has moved, 2016 → 2026
The Interplanetary Transport System
At the IAC in Guadalajara Musk unveils a carbon-fibre Mars rocket. Earlier that year SpaceX talked of a Red Dragon capsule landing on Mars by 2018.
Cargo landings targeted for 2022
At the IAC in Adelaide the vehicle shrinks and is renamed BFR; the first cargo ships are now pencilled for the 2022 window.
Five uncrewed ships in 2026, crew in 2028
Musk sets the plan the world remembers: five uncrewed Starships in the 2026 window; if they land safely, humans in 2028; the fleet grows exponentially each window.
Optimus will go first
The 2026 ships will carry Tesla Optimus robots; humans “as soon as 2029, although 2031 is more likely.”
“50/50” — The Road to Making Life Multiplanetary
In a Starbase talk Musk lays out the full plan (below) and puts the odds of a 2026 launch at 50/50, naming in-orbit refilling as the biggest hurdle.
Sours on 2026
Musk posts that the first flight without humans is ~3.5 years away and the first with humans ~5.5 years — i.e., 2028–29 and 2030–31 — with only a “slight chance” of anything in late 2026.
Moon first
SpaceX tells investors the Moon comes first; Musk says a self-growing lunar city within a decade and a Mars city effort in five to seven years.
Starship reaches orbit
The prerequisite finally met — nine weeks before the Mars window, and without refueling.
Window opens, nothing launches
The next low-energy window is late 2028 to early 2029.
The plan itself: what Musk laid out on May 27, 2025
The most detailed public Mars plan remains the Starbase talk “The Road to Making Life Multiplanetary.” It is still the reference, even though its first step has slipped a window. Space.com
| Window | Ships | Purpose |
|---|---|---|
| 2026 | 5 uncrewed V3 ships | Carry Tesla Optimus robots; prove landing. Super Heavy stays on Earth. |
| 2028–29 | ~20 ships | More Optimus to build infrastructure and survey water ice; possibly humans on one ship, or two more robot rounds first. |
| 2030–31 | ~100 ships | Begin the buildout; crewed ships carry 100–200 passengers each. |
| 2033 | ~500 ships | Scaling. |
| Mature | 1,000–2,000+ ships per window | A self-sufficient city of more than a million people needing millions of tonnes of cargo. |
- Landing site: Arcadia Planitia, a flat northern lowland with accessible water ice, away from the poles.
- First ship's name: Heart of Gold, after the Hitchhiker's Guide starship.
- Vehicle: V3 at 124.4 m; a future 142 m variant was mentioned for the city buildout.
- Transit: with a full propellant load, Earth to Mars in roughly 3–4 months.
Progress is measured by the timeline to establishing a self-sustaining civilization on Mars.Elon Musk, Starbase, May 27, 2025
SpaceX's own Mars vision film. The May 2025 talk is on the video page.
The hard part: refueling in orbit
A Starship arriving in low Earth orbit has nearly empty tanks. To reach Mars it must be refilled — roughly 1,200 tonnes of propellant (about 933 t liquid oxygen, 267 t methane). With roughly 100 t delivered per tanker flight, that is about a dozen tanker launches per Mars ship. Five ships in 2026 would have needed around 60 tanker flights. Aerospace America
What has been done: on Flight 3 (March 2024) SpaceX moved roughly 10 tonnes of oxygen between two tanks inside one ship, a NASA Tipping Point milestone. It remains the only on-orbit transfer test to date. The ship-to-ship demonstration was due around March 2025, then March 2026; Shotwell now targets early 2027.
On Mars, the return trip depends on making propellant locally — solar power plus the Sabatier reaction turning CO₂ and water ice into methane. One analysis estimates ~38,400 m² of solar panels and ~33 tonnes of regolith processed per day for 5 t/day of propellant. For perspective, NASA's MOXIE experiment made 122 grams of oxygen in 30 months.
Robots before people
Since March 2025 the plan has been for the first uncrewed ships to carry Tesla's Optimus humanoid robots to install power and infrastructure and scout resources before any crew arrives. Musk at the May 2025 talk: “That would be an epic picture — to see Optimus walking around on the surface of Mars.”
In October 2026 Musk named robotics as a reason to merge SpaceX and Tesla, which would bring Optimus in-house. Experts quoted by Aerospace America doubted a humanoid is the right first tool and suggested rovers or helicopters. Optimus's own progress is tracked on Tesla Unpacked.
Scaling toward a fleet
- Starfactory at Starbase targets one ship per day at peak.
- Starbase Louisiana — $100B, 12 to 20+ towers, “thousands of flights a year,” first launch 2028–29.
- Cadence targets — weekly flights in H1 2027, twice weekly by end-2027; Musk's aspiration is “well over 1 million tons to orbit per year” and more than 10,000 launches a year by 2030.
- The incentive — Musk's IPO share grant includes 1 billion performance shares that vest only on establishing a permanent Mars colony of at least 1 million inhabitants. Fortune / S-1
When do people go?
The best current reading of Musk's statements: an uncrewed landing attempt in the 2028–29 window, crew in 2030–31, and the beginning of a city effort around 2031–33. In August 2025 he said a self-sustaining city is “20 to 30 years” away. Every one of these dates has moved before, and the company's stated priority through 2028 is the Moon.
The planned first crewed Starship flight, Polaris III, was to be commanded by Jared Isaacman after roughly 100 successful cargo flights — but Isaacman now runs NASA, so the crew is an open question.