Nvidia Launches Model Revolutionizing Physical AI

Nvidia unveiled a new set of AI technologies designed to advance physical AI. These technologies target autonomous vehicles and robotics capable of perceiving and interacting with the real world.

Central to this announcement is Alpamayo-R1, an open vision-language reasoning model for autonomous driving research, presented at the NeurIPS AI conference in San Diego. This model integrates visual and textual inputs, allowing vehicles to understand their environment and make informed decisions.

Alpamayo-R1 builds on the Cosmos-Reason framework, a reasoning model that evaluates options before acting. Originally launched in January 2025, the Cosmos model family received additional updates in August. Nvidia highlights that this technology is key for achieving level-4 autonomy. At this level, vehicles operate independently within defined areas and conditions, using AI capable of human-like judgment.

In tandem, Nvidia released the Cosmos Cookbook on GitHub. It provides developers with detailed workflows, data curation guidance, synthetic data generation tools, and evaluation methods. These resources facilitate the efficient training and adaptation of Cosmos models for specialized tasks in robotics and autonomous driving.

Routine tasks take a lot of time from employees? Do they burn out, do not have enough working day for the main duties and important things? Do you understand that the only way out of this situation in modern realities is automation? Try Apix-Drive for free and make sure that the online connector in 5 minutes of setting up integration will remove a significant part of the routine from your life and free up time for you and your employees.