Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Nvidia announced today that it is releasing a family of basic AI models called Cosmos that can be used in training. humanoidsindustrial robots, and self-driving cars. While language learners learn to create text by teaching books, articles, and TV shows, Cosmos is designed to create images and 3D models of the world.
In a keynote speech at the annual CES conference in Las Vegas, the CEO of Nvidia Jensen Huang showed cosmos models used to simulate scenes inside warehouses. Cosmos was trained on 20 million hours of real-time footage of “people walking, hands moving, manipulating things,” Jensen said. “It’s not about creating artificial things, it’s about training AI to understand the environment.”
Researchers and inventors hope that these basic types can provide robots used in factories and homes the highest level of skill. For example, cosmos can create video boxes falling from shelves inside a warehouse, which can be used to train a robot to recognize danger. Users can also fine-tune the models using their own data.
Several companies are already using Cosmos, Nvidia says, including humanoid robots Agility and Image AI and self-driving companies like Uber, Waabi, and Wayve.
Nvidia also announced software designed to help different types of robots learn to perform new tasks better. The new feature is part of Nvidia’s existing Isaac robot simulation platform that will allow robot builders to take a few samples of a desired task, such as handling an object, and generate large amounts of training data.
Nvidia hopes that Cosmos and Isaac will attract companies that want to create and use humanoid robots. Jensen was joined on stage at CES by giant images of 14 different robots made by companies including Tesla, Boston Dynamics, Agility, and Image.
Along with Cosmos, Nvidia also announced Project Digits, $3,000 “personal AI supercomputer” which can run up to 200 billion languages without the need for cloud services from the likes of AWS or Microsoft. It also announced the much-anticipated RTX Blackwell GPUs, and upcoming hardware to support production. AI assistants.
