NVIDIA DRIVE Hyperion

Driving Next-Generation AV Breakthroughs The future of mobility is driven by AI, and NVIDIA’s in-vehicle compute solutions are central to this transformation. The [NVIDIA DRIVE Hyperion™](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/drive-hyperion/) platform features two [NVIDIA DRIVE AGX Thor™](https://developer.nvidia.com/drive/agx) systems-on-a-chip on a single board, the safety-certified [NVIDIA DriveOS™](https://developer.nvidia.com/drive/os) operating system, and a fully qualified multimodal sensor suite that includes 14 high-definition cameras, nine radars, one lidar, and 12 ultrasonics. This delivers the compute performance and redundancy required for [L4 autonomous driving](https://blogs.nvidia.com/blog/level-4-autonomous-driving-ai/). At its core, NVIDIA DRIVE AGX™ provides the AI computing power for safe, intelligent driving, while [NVIDIA DRIVE AV](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/drive-av/) delivers the end-to-end [autonomous driving](https://www.nvidia.com/en-in/glossary/autonomous-vehicles/) software stack. By integrating the [NVIDIA Alpamayo](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/alpamayo/) family of open-source models, tools, and data for reasoning-based autonomy, the platform helps vehicles solve complex, long-tail scenarios with human-like judgment and transparent decision-making. Together with the [NVIDIA DriveOS](https://developer.nvidia.com/drive/os) safety-certified operating system, this enables real-time processing, security, and system monitoring to meet functional safety requirements.

Unmatched AI Computing for Autonomous Vehicle Innovation NVIDIA DRIVE AGX delivers a scalable, energy-efficient, future-proof AI computing platform, engineered to process the most complex, real-time workloads essential for safe autonomous driving. End-to-End Safety and Security Safety-certified operating system (DriveOS), and hardware design principles comply with stringent automotive safety standards, providing a foundation of trust. NVIDIA Halos provides deployment-time guardrails for run-time monitoring and real-time introspection, ensuring continuous protection and security across the entire AV stack, from the cloud to the car. Future-Proof AI-Defined Architecture Runs the latest and most demanding AI models, ensuring your vehicle’s capabilities evolve throughout its entire lifecycle with support for continuous hardware and software upgrades. Sensor Fusion Integrates data from the complete suite of sensors—including cameras, radar, lidar, and ultrasonic—to create a robust, 360-degree real-time perception of the environment. Optimized Energy Efficiency Features a power-efficient, in-vehicle architecture that delivers maximum AI performance with minimal thermal and energy impact.

A Powerful AI Computing Platform Delivering scalable, high-performance hardware for highly automated and fully autonomous driving. NVIDIA DRIVE Hyperion - Two DRIVE AGX Thor in-vehicle computers based on Blackwell architecture. - Features 14 high-definition cameras, 9 radars, 1 lidar, and 12 ultrasonic sensors and a microphone array. - Includes DRIVE™ AV software, purpose-built for L4 autonomy. NVIDIA DRIVE AGX ThorTM - Delivers more than 1,000 INT8 TOPS (2,000 FP4 FLOPs). - Offers a scalable architecture supporting Level 2+ to fully autonomous driving. - Provides superior safety, with ASIL-D compliance and redundancy. NVIDIA DRIVE AGX OrinTM - Delivers up to 254 TOPS of AI performance. - Offers a scalable architecture supporting Level 2+ to fully autonomous driving. - Provides superior safety, with ASIL-D compliance and redundancy.

Breakthroughs for Autonomous Vehicles Enabling safe, scalable, and intelligent driving systems. NVIDIA Alpamayo - Open and Transparent AI: A portfolio of open AI models, simulation frameworks, and AI datasets that teams can inspect and adapt for transparent, reproducible AV development. - Explainable Long‑Tail Autonomy: Vision‑language‑action models that reason through complex scenarios and verbalize decisions, making driving behavior interpretable and auditable. - An Accelerated [Path to Level 4:](https://www.youtube.com/watch?v=mYUrtm1W-Mg&t=2s) High‑fidelity, closed‑loop simulation that shifts miles from road to lab, reducing test time and accelerating validation. NVIDIA DriveOS - ASIL-D Safety Foundation: A production-proven, certified architecture that provides the base for the most advanced, mission-critical applications. - Unified AI Acceleration: Seamlessly powers both high-performance autonomous driving and an immersive cabin AI experience through dedicated acceleration for [LLMs](https://www.nvidia.com/en-in/glossary/large-language-models/) and [VLMs](https://www.nvidia.com/en-us/glossary/vision-language-models/). - Unparalleled Versatility for Software Deployment: Cloud-to-car deployment for both QNX and Linux while providing a holistic security framework for the era of Software Defined Vehicles. NVIDIA DRIVE AV - Features full-stack software that includes active safety, parking, and hands-off and eyes-off driving. - Provides state-of-the-art software architecture using an end-to-end model for self driving that ensures natural driving behavior, even in complex scenarios. - Features the [NVIDIA Halos](https://www.nvidia.com/en-in/self-driving-cars/safety/) safety guardrail system that prioritizes both vehicle and occupant protection, from the cloud to car.

Accelerated Training and Inference for Autonomous Vehicle Development A suite of optimized AI models now available for AV developers. bevformer Advanced transformer for multi-frame, bird’s-eye-view 3D perception in autonomous driving. sparsedrive End-to-end autonomous driving stack integrating perception, prediction, and planning with sparse scene representations for efficiency and safety. streampetr Efficient 3D object detection for autonomous driving by propagating sparse object queries temporally.

A Shared Vision With Industry Leaders Delivering industry-leading platforms for the world’s top automakers and mobility companies, built on NVIDIA DRIVE AGX. Mercedes-Benz NVIDIA is enabling a new era of AI-defined driving, bringing its level 2++ capabilities to U.S. roads this year, starting with Mercedes-Benz, a long-standing partner in advancing safe, intelligent mobility. Toyota Toyota, the world’s largest automaker, will build its next-generation vehicles on [NVIDIA DRIVE AGX OrinTM](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/in-vehicle-computing/), running the safety-certified NVIDIA DriveOS operating system. General Motors GM, the largest U.S. automaker, will use NVIDIA AI, Omniverse and DRIVE AGX to develop and build next-gen vehicles, factories, and robots. JLR Starting in 2026, JLR vehicles will be built on the [NVIDIA DRIVE](https://www.nvidia.com/en-in/self-driving-cars/) AI-defined platform—from the cloud, built on NVIDIA data center solutions, to the car, powered by [NVIDIA DRIVE AGX Orin](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/in-vehicle-computing/). Volvo Cars The Volvo EX90 is built on the [NVIDIA DRIVE AGX Orin](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/in-vehicle-computing/) system-on-a-chip (SoC), with plans to migrate to the NVIDIA DRIVE Thor computer, running [DriveOS](https://www.nvidia.com/en-in/solutions/autonomous-vehicles/in-vehicle-computing/#software) for its future fleet of cars.