Nvidia Opens an Autonomous-Driving Model The model is Nvidia Alpamayo 2 Super, a 34-billion-parameter multi-task foundation model for autonomous-vehicle development. It combines a 32-billion-parameter vision-language backbone with a 2-billion-parameter diffusion action expert. It can process six-camera sequences, produce reasoning traces and sample future trajectories. It can be downloaded from the NVlabs GitHub repository. The model…
To access this post, you must register DVN Sensing & Applications Gold Membership or DVN Sensing & Application Platinum Membership.
