Diligent Robotics, a subsidiary of Serve Robotics (Nasdaq: SERV), has begun deploying its Moxi 2.0 robot platform to health system customers across the U.S., according to a company statement.

The updated platform features upgraded Nvidia-powered A2000 compute that the company says processes perception data 10 to 15 times faster than the previous generation. Moxi 2.0 was developed using Nvidia Isaac Sim and aspects of the Nvidia Cosmos open world models. The robot can operate up to nine hours on a single charge, with 30% faster charging than before, enabling up to 18 hours of daily operation.

The rollout includes deployments at Endeavor Health Edward Hospital, Providence Saint John's Health Center, and Children's Hospital Los Angeles. Omkar Kulkarni, Chief Innovation and Transformation Officer at Children's Hospital Los Angeles, said Moxi has completed more than 40,000 deliveries at the facility, accounting for over 16,000 hours of work, and that the hospital expanded its Moxi fleet from two to three robots, with utilization rising more than 10% in the second quarter.

Diligent said the platform includes a proprietary World Model that uses fleet deployment data to improve robot software over time. Cloud infrastructure supporting the system was built with Amazon Web Services, with model training conducted on Amazon SageMaker HyperPod. T-Mobile for Business collaborated with Diligent on network connectivity, including planning for private 5G hospital deployments.

"Hospital hallways are some of the most dynamic environments a robot can operate in," said Andrea Thomaz, founder and CEO of Diligent Robotics. "Moxi 2.0 is built with the compute, sensors and models to reason about that complexity in real time, so Moxi can move with more confidence without compromising safety."

The Moxi platform operates without requiring facility modifications or IT infrastructure changes. Diligent said it is accepting orders from additional health systems. The company, founded in 2017 and based in Austin, Texas, was acquired by Serve Robotics in 2026.