Agility releases new version of their humanoid robot

Agility, the homegrown humanoid robotics startup that came out of Oregon State, has released a new version of its robot. Meet Digit 5.

Built upon a decade of research and years of deployment experience, our fifth generation humanoid is the most powerful automation tool we’ve ever built. Engineered for cooperative safety, designed to work in close proximity to people, adaptable to handle end-to-end workflows.

“Digit 5 is removing a major barrier to scaling humanoid robots in industrial environments,” said Peggy Johnson, CEO of Agility Robotics. “We built it to the exact requirements our customers gave us after three years of Digit 4 working on their production floors. The market has responded with more than $300 million in multi-year orders and a growing pipeline of customers across manufacturing, warehousing and logistics. That’s value that compounds as Digit’s capabilities expand, backed by real demand.”

The upgrades…?

• 40% more payload — A new leg design engineered to withstand repetitive lifting, powered by proprietary cycloidal actuator technology, lets Digit 5 repeatedly lift up to 50 lb. (22.7 kg) loads, allowing full coverage of single-person lift tasks in OSHA-regulated facilities.
• Faster autonomous charging for more time working — A new 90-minute runtime battery system charges in just 9 minutes, giving Digit 5 a 10:1 run-to-charge ratio (up from Digit 4’s 2:1), enabling more than 20 hours of productive work in a 24-hour day.
• Swappable end-effectors — A swappable gripper design with ISO-standard mounting flanges allows Digit’s tools to be changed for the job at hand, enabling greater productivity while tackling tasks ranging from tote handling to machine tending.
• Higher reach with a human-scale footprint — Digit 5 weighs 284 lb. (129 kg), stands 5’ 11” (1.81 m), and can reach heights of up to 7.2 feet (2.2 m) (up from Digit 4’s 5.5 ft), sized to reach the same shelves and move through the same aisles, doorways and workstations built for people.

And while there’s clearly a ton of changes to the device — and likely some huge leaps forward in terms of physical AI — what strikes me most about this version is how much more stocky it is. Especially when compared to early more bird like versions of the robot. (And I kinda miss the backwards bending knees.)

Speaking of robots. This piece about physical AI manifested in robots landed in a timely fashion.

With LLMs, the hardware bends to the model. Pour in as much data and compute as possible at training, then figure out how to serve the model that comes out. The model comes first; the hardware second. Robotics inverts this due to two constraints.

The first is time. A robot runs real-time control loops and can’t miss a deadline. An LLM can be slow without affecting the final output. If a robot is slow, the world changes around it, and then the action is obsolete by the time it needs to enact it.

The second is cost. When you use an LLM, it lives behind a screen provided by the user. In robotics, the manufacturer has to build both the compute and the robot itself and pay for it upfront, on every unit, and the capital expenditure can be immense. At scale, this upfront cost can be billions, even in the trillions if we do eventually get to a billion robots.

For more on this release, read the press release or visit the Digit 5 page at Agility.