NVIDIA CEO Jensen Huang given a new leather jacket by a humanoid robot from 1X Technologies

We all know that the leather jacket that NVIDIA founder and CEO Jensen Huang wears is iconic, to say the least, but now he has a new one: gifted to him by a humanoid robot made by 1X Technologies.

Jensen recently visited the 1X Technologies HQ in San Francisco, where the NVIDIA CEO was greeted by a humanoid robot, gifting him one of the very best black leather jackets on the market, custom-made by California-based ERL clothing brand. 1X Technologies might be a name you haven't heard of, but the company has been working on its Neo Gamma humanoid robot, with plans to ship hundreds of thousands of them before the end of 2025.

As for Jensen's new leather jacket, but sports an NVIDIA logo embroidered into the clothing, and I'm sure will join the wardrobe of leather jackets that the NVIDIA CEO has in his house. It looks awesome, and I definitely want one.

1X Technologies explains on its website: To make this collaboration possible, the 1X AI Team created a dataset API for NVIDIA to access data collected from 1X offices and employee homes, and an inference SDK to serve model predictions at a continuous 5Hz vision-action loop using an onboard NVIDIA GPU in NEO's head or an offboard GPU.

3

A crucial step when onboarding a new learning codebase onto NEO is to verify correctness, i.e., overfitting a baseline model to a small amount of demonstration data and making sure that the time synchronization between images and actions is consistent all the way from data collection to training to runtime inference.

We demonstrate this by working with the NVIDIA GEAR team to train a single end-to-end neural network based on the NVIDIA GR00T N1 model to autonomously grasp a cup, hand it over to the other hand, and place it in a dishwasher to showcase how NEO fits compactly into the kitchen space while still having the kinematic reach to carry the cup from sink to dishwasher.

This is a good "first task" to learn because it checks for basic compatibility of an external research codebase with the logging and inference architecture. The obvious next step after verifying correctness is to feed thousands of hours of internally collected NEO data into the model.

3

Over the course of a week, our teams developed this model at a 1X employee's home, swapped notes on action spaces, control frequencies, and other imitation learning tricks needed to get good performance on NEO Gamma. Moments like these - where friends are just hanging out in the home while a NEO does dishes in the background - will soon become an everyday occurrence.