Agility Robotics Powers the Future of Robotics with NVIDIA

A significant step forward in our deep technical collaboration.

Agility Robotics Powers the Future of Robotics with NVIDIA
Published on
September 10, 2025

At Agility Robotics, our mission has always been to build robots that make a positive impact on the way we work and live — machines that can move, work, and adapt alongside humans in dynamic environments. From the very beginning, we’ve known that achieving this vision requires not only cutting-edge hardware, but also the most advanced computing platforms available.

That’s why our relationship with NVIDIA has been such an important part of our journey. You may have seen Agility with the NVIDIA team at GTC the past few years or at CES as part of the NVIDIA keynote this past January. In fact, one of our earliest demonstrations of using LLMs to control Digit was done at GTC 2024.

Over the years, NVIDIA has helped us bring breakthrough AI and autonomy to life in our bipedal humanoid robots, from the earliest prototypes to the highly capable systems we’re deploying today.

A Common Commitment to Innovation

When Agility Robotics set out to design robots like Digit, we were tackling challenges at the intersection of locomotion, perception, and intelligence. NVIDIA AI infrastructure and developer frameworks made it possible for us to train robust models and process real-world sensor data in real time. At the heart of Digit’s onboard compute is an NVIDIA accelerator, one that has been evolving from NVIDIA RTX hardware in earlier versions of Digit toward the purpose-built NVIDIA Jetson AGX Thor robotics computer. Obtaining early access to Jetson Thor is an example of our continued collaboration and an exciting opportunity to advance robotic intelligence together.

By integrating NVIDIA’s platforms with our expertise in robot design and control, we’ve been able to push the boundaries of what robots can do — from navigating complex spaces, to responding to unexpected changes, to working seamlessly in human-centered environments.

Enabling Scalable AI for Robotics

As robotics moves into the next phase —large-scale deployment in warehouses and factories and beyond to homes, hospitals, hotels, offices, and retail stores — NVIDIA’s technology is a cornerstone in helping us scale. Their platforms give us the computing performance and flexibility to:

  • Train perception and control models more quickly and accurately with scalable NVIDIA GPU training in the cloud.
  • Run sophisticated AI algorithms at the edge and onboard the robot, where real-time responsiveness is critical.
  • Leverage high-fidelity simulation environments to safely develop, test and refine behaviors before real-world deployment.

The path to deploying robots at scale requires training new skills in simulation and testing and refining those skills by deploying them to real hardware in the field. Robotic simulation frameworks such as NVIDIA Isaac Lab make that sim-to-real transfer possible.

This kind of performance has been instrumental in making Digit not just a demonstration of what’s possible, but a practical and reliable co-worker.

Looking Ahead Together

Our ongoing relationship with NVIDIA is more than a technology collaboration — it’s a commitment to building the future of robotics. With NVIDIA’s support, we’re unlocking new ways for intelligent machines to assist people in industries that need them most.

A natural evolution of our mutual commitment to the future of robotics, we received an investment from NVentures (NVIDIA’s venture capital arm) in our Series C round.

As we continue scaling Digit deployments and expanding into new applications, we’re excited to keep collaborating withNVIDIA on the next generation of AI-accelerated robotics. The road ahead is full of opportunities, and with NVIDIA’s innovation alongside ours, we’re confident that the future of robotics will be more capable, more adaptive, and more human-centered than ever before.

At Agility Robotics, we’re proud of how far we’ve come with NVIDIA’s support — and even more excited about where we’re going next.

The Agility newsletter
No spam. Just occasional updates, and exclusive interviews from the front lines of building the world's most advanced humanoid robot.
Read about our privacy policy.
Thank you! You will get the latest updates straight to your inbox.
Oops! Something went wrong while submitting the form.

Agility in the press.