Accelerating the Future of Autonomous Vehicles
Pioneering Safety in Autonomous Vehicles (AVs)
NVIDIA is redefining autonomous vehicles (AV) with its three-computer solution. AI models are trained on NVIDIA DGX™ to tackle complex driving scenarios, while NVIDIA Omniverse™ and Cosmos™ simulate diverse physically-based environments to test and validate AV systems. In the vehicle, NVIDIA DRIVE AGX™ enables safe, real-time decision-making on the road. NVIDIA Halos extends this safety-first approach by providing guardrails across simulation, training, and deployment, setting a new industry-leading standard.
AI Training at Scale
NVIDIA DGX delivers a high-performance AI training compute platform designed to accelerate autonomous vehicle development. Developers can further reduce bottlenecks with NVIDIA Cosmos data curation services to manage the massive datasets necessary for end-to-end training. Combining best-in-class NVIDIA software, infrastructure, and expertise in a unified AI development solution lets AV developers train and scale AV systems more efficiently.
In-Vehicle Compute Platform
NVIDIA DRIVE AGX Hyperion™ 10 is a reference production computer that streamlines development for autonomous systems. Featuring two NVIDIA DRIVE AGX Thor™ systems-on-a-chip on a single board; the safety-certified DriveOS™ operating system; and a fully qualified multimodal sensor suite that includes 14 high-definition cameras, 9 radars, 1 lidar, and 12 ultrasonics and a microphone array— it delivers the compute performance and redundancy necessary to enable reasoning VLA end-to-end network models required for L4 autonomous driving.
Building Safe, AI-Driven Autonomous Vehicles
The Path to Level 4 Autonomy
Recent breakthroughs in AI—from foundation and reasoning models to high-fidelity simulation and safety validation—are making level 4 autonomy a reality. NVIDIA’s end-to-end platform forms a rich ecosystem for learning, testing, and deployment, accelerating innovation while keeping safety at the center.
NVIDIA Halos: A Full-Stack Comprehensive Safety System for Autonomous Vehicles
Over 15,000 engineering years invested in AV safety have led to the NVIDIA Halos system for chip-to-deployment AV safety. It combines hardware, software, tools, models, and proven design principles to safeguard end-to-end AV stacks.
Autonomous Vehicle Innovation In Action
Turbocharge Autonomous Vehicle Development with
NVIDIA Inference Microservices (NIM™)
Tap into NVIDIA advanced AI models for AV-embedded software and cloud solutions to accelerate real-world deployment and improve efficiency.
Latest Keynotes and Demos
NVIDIA Keynote at GTC DC 2025
NVIDIA Keynote at GTC 2025
Accelerating AV Development with Omniverse and Cosmos
NVIDIA Halos - A Comprehensive Safety Systems for Autonomous Vehicle Development
NVIDIA CES 2025 Automotive Recap
NVIDIA Cosmos: A World Foundation Model Platform for Physical AI
Scaling AV Data With Omniverse and Cosmos