DRIVE Thor
System-on-chip for autonomous vehicles and robotaxis
NVIDIA DRIVE Thor is a system-on-chip designed for autonomous vehicles and robotaxis. It serves as the compute core of the NVIDIA Hyperion platform, enabling Level 4 autonomy through redundant processing and safety mechanisms.
- Targets Level 4 autonomous vehicle capabilities
- Part of the NVIDIA Hyperion sensor and compute platform
- Meets ISO 26262 ASIL-D safety requirements
- Supports redundant compute for enhanced safety
What it is
DRIVE Thor is a system-on-chip (SoC) developed by Nvidia for use in autonomous vehicles. It acts as the central computing unit for self-driving cars and robotaxis. The chip supports high levels of automation, specifically targeting Level 4 capabilities. It is part of the broader NVIDIA Hyperion platform, which combines sensors and compute hardware. This setup allows developers to build safer autonomous systems with reduced integration risks.
Inside the chip
The chip features redundant compute capabilities to ensure functional safety. It meets ISO 26262 ASIL-D requirements, a standard for automotive safety integrity. DRIVE Thor works alongside a standardized sensor suite that includes cameras, radar, lidar, and ultrasonics. This architecture allows the vehicle to process data from multiple sources simultaneously. The design supports over-the-air updates, enabling new features to be added without changing hardware.
Where it went
This processor is used in passenger vehicles, robotaxis, shuttles, and delivery fleets. It powers the NVIDIA Hyperion platform, which is adopted by various mobility leaders in Europe and globally. The chip enables the deployment of reasoning-based AI models like NVIDIA Alpamayo. These models help vehicles navigate complex real-world scenarios by understanding cause and effect. It supports the full lifecycle of autonomous driving development, from simulation to road deployment.
Versions and family
DRIVE Thor is positioned as the next generation following the NVIDIA DRIVE Orin. Both chips are part of the NVIDIA DRIVE AGX family. The platform scales from Level 2++ driver assistance up to Level 4 autonomy using the same underlying architecture. Higher levels of automation are unlocked via software updates rather than new hardware installations.
Questions people ask about DRIVE Thor
What is NVIDIA DRIVE Thor used for?
It is a system-on-chip used in autonomous vehicles and robotaxis. It provides the computing power needed for Level 4 self-driving capabilities and processes data from various sensors.
How does it ensure safety?
The chip includes redundant compute systems and meets ISO 26262 ASIL-D safety standards. It works with the NVIDIA Halos system to validate safety across the entire development lifecycle.
Can it be updated after purchase?
Yes, the platform is software-defined and capable of over-the-air updates. This allows vehicles to gain new autonomous features and adapt to regulations throughout their lifetime.
Good reading and sources
There is no English Wikipedia article on DRIVE Thor yet, so the facts on this page come from the sources listed above: nvidia.com/en-us/self-driving-cars/in-vehicle-computing/. Some are the subject's own pages, so its claims about itself are reported as its claims. EveryChip wrote the text in its own words, with AI assistance, and compares it with the sources to make sure no passage is copied.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.