At a glance
Agility Robotics announced Digit 5 on September 15, 2026, describing a humanoid built for work near people and warehouse workflows. Its safety and performance figures are company claims, and the announcement does not provide independent Digit 5 test results. Agility Robotics
Digit 5 announcement centers on working near people
Agility Robotics announced Digit 5, the next generation of its general-purpose humanoid robot, on September 15, 2026. The company presents the model as intended to work close to people without the physical safety barriers used in traditional automation, making collaborative operation a central part of its pitch. Agility Robotics
Agility describes a safety architecture combining human detection, visual and auditory cues, and an independent safety controller. Those are design claims from the manufacturer, rather than independently reported test outcomes, so the announcement establishes what the company says it has built but not how the system performs across varied work sites. Digit 5’s safety architecture
Charging, payload and workflow integration shape the operating case
The company says Digit 5 can repeatedly lift loads up to 50 pounds, or 22.7 kilograms, and lists a 90-minute runtime battery that charges in nine minutes. Agility describes this as a 10:1 run-to-charge ratio and says the schedule could support more than 20 hours of productive work within a 24-hour day, not continuous operation on one battery. Agility’s charging and runtime claims
Agility also lists swappable grippers with ISO-standard mounting flanges and says workflows can include depalletizing, machine tending, kitting, sequencing, inspection and palletizing. The company says its robot-management platform coordinates handoffs with autonomous mobile robots, conveyors and warehouse systems, placing integration alongside the robot’s physical capabilities in the proposed use case. Agility Robotics
Editorial analysis: APAC buyers should test the operating assumptions
For APAC operators, the useful question is not whether a humanoid can be shown performing a task, but whether its safety and work cycle fit a particular facility’s people, shifts and material flow. The announcement’s combination of close-proximity safety claims, quick charging and handoffs with existing systems gives buyers concrete areas to evaluate, without establishing that the same configuration suits every site.
A practical assessment should ask how human detection and the independent safety controller are validated around workers, what visual or auditory cues communicate in a noisy facility, and how interventions affect task completion. Buyers should also check whether the claimed lifting capacity and listed workflows match their actual loads, grasp requirements and process tolerances, rather than treating capability descriptions as demonstrated production output.
Field record, availability and evidence still require scrutiny
Agility reports more than 65,000 hours of Digit 4 operational time across customer sites in North America, and says Digit 4 reached a cumulative 100,000-tote milestone at GXO’s Flowery Branch facility at approximately 98% accuracy while on task. These are company-reported results for Digit 4, not independent Digit 5 test results, and the announcement does not establish that the two models share identical performance. Agility’s Digit 4 operating record
The company expects early access to Digit 5 to begin in the first half of 2027 and general availability by the end of 2027; both are plans, not completed milestones. Before treating the product as ready for deployment, readers should seek independent safety validation, measured task performance in relevant conditions, the basis for the runtime and charging figures, and evidence of how planned integrations work in customer operations. Agility Robotics
Sources and reporting notes
Based on Agility Robotics’ company announcement, not an independent media account. Safety, specifications and performance statements are company claims; independent Digit 5 test results are not provided in the announcement.
Original synthesis and analysis of the linked public reporting. AI-assisted drafting and translation; reported claims remain attributed to their sources.