Agility Robotics Explores Wheels for Digit, Its $140 Million Loss

Hurst says wheels are 'too early to say' on timing

Jonathan Hurst, Agility Robotics' co-founder and chief robot officer, said Sept. 25 that the company is exploring wheeled versions of its Digit humanoid alongside the legged design it has sold since 2023. "We're exploring a range of robot designs and form factors, including wheels, to address different operating environments and customer needs," Hurst told The Robot Report by email, after a wheeled robot rendering appeared briefly near the end of the Digit 5 launch video on Sept. 15.

Asked when a wheeled robot might reach customers, Hurst gave no date: "Too early to say! ... I have guesses and inclinations, but really we're just starting." Agility has not said whether a wheeled machine would be a Digit variant riding on a base instead of legs, or a separate product line entirely.

The $300 million order book comes from one customer

The wheels comment landed ten days after Agility's own numbers became public for the first time. Agility is going public through a $2.5 billion merger with Churchill Capital Corp. XI, and the S-4 filing tied to that deal disclosed Digit's finances: $1.8 million in 2025 revenue against a $140 million operating loss. The same filing states that Agility has more than $300 million in multi-year Digit v5 orders, and The Robot Report reported that the order book comes from one customer.

One customer, not a market. A logistics operator counting on a single supplier for $300 million of future robot deliveries is a large bet on Agility specifically, and Agility's own numbers show a company still years from recognizing revenue anywhere near that order backlog: $1.8 million booked against $300 million promised is roughly half a percent realized so far. That concentration cuts both ways. If the one customer expands its order, Agility's revenue line moves fast off a near-zero base; if that customer pauses or renegotiates, the backlog figure that anchors the SPAC valuation moves just as fast in the other direction, and there is no second contract disclosed in the filing to absorb the shock.

Agility is not alone in pairing large backlog numbers with small realized revenue. It is the standard shape of a company selling a capital good that customers pilot for months before committing to volume, and the S-4 process exists specifically so a public market, rather than only venture investors, gets to see that shape before buying in.

Digit 5's safety pitch, priced against 65,000 hours of runtime

Agility's Sept. 15 launch material for Digit 5 says the robot has logged more than 65,000 hours of deployment and hit a 100,000-tote milestone at a GXO logistics facility, its first cited customer site. Chief Executive Peggy Johnson said in that release that "Digit 5 is removing a major barrier to scaling humanoid robots in industrial environments," pointing to a new leg design meant to withstand repetitive lifting without the safety cages earlier industrial robots required.

Metric	Figure
2025 revenue recognized	$1.8 million
2025 operating loss	$140 million
Multi-year order book	$300+ million, one customer
Logged deployment hours	65,000+
Payload capacity	50 lb, up 40% over Digit 4
Runtime per charge	90 minutes

Digit 5's payload rose to 50 lb, a 40 percent increase over Digit 4, with a 90-minute runtime and a nine-minute recharge, a 10-to-1 run-to-charge ratio Agility says covers single-person lift tasks under OSHA rules. Deepu Talla, Nvidia's vice president of robotics and edge AI, is quoted in the same release saying safety around people and assets is what moves humanoid robots "from pilots to industrial deployments at scale," a claim about Nvidia's own Halos safety platform, which Digit 5 uses, rather than an independent test of Digit itself.

Legged robots working in warehouses have limits a robot on wheels would not

The tire-changing robot PitPro built for a Calgary Kal Tire already runs on wheels rather than legs, for the plain reason that a wheeled base is cheaper, more stable and faster over flat concrete than a bipedal one, and legs only earn their cost where stairs, uneven ground or human-scale doorways make wheels impractical. Digit's legs were the pitch precisely because a warehouse has ramps, pallets and clutter a wheeled robot cannot cross, which is what makes Hurst's wheels comment notable: it concedes that some of Digit's own customer environments may not need legs at all, and that a wheeled competitor could undercut Digit on cost in exactly those sites.

That is also the logic behind Amazon's newly announced $100 million robotics manufacturing plant in Indiana, which is built around wheeled and fixed automation rather than humanoids, betting that most warehouse tasks do not need a machine shaped like a person.

What would change this read

Agility's Analyst and Investor Day is set for Oct. 6, the first scheduled moment for the company to say more about the wheeled concept, the concentration risk in its one-customer order book, or a timeline for turning $300 million in orders into recognized revenue. If Agility names a second major customer at that event, the concentration risk shrinks. If it does not, and Hurst's wheeled robot never gets a launch date, the wheels comment reads less like a product roadmap than an answer given because a journalist noticed a frame in a launch video.