Build the Robots of Tomorrow with NVIDIA Jetson Thor
NVIDIA has unveiled the NVIDIA® Jetson Thor™ series, a groundbreaking platform for physical AI and robotics. Built on the NVIDIA Blackwell architecture, Jetson Thor delivers a massive leap in on-device AI compute, engineered to meet the demands of next-generation autonomous machines.
With up to 2,070 FP4 teraflops of AI performance, 128 GB of memory, and configurable power from 40 W to 130 W, Jetson Thor is built for real-time robotic applications that require high-throughput sensor fusion, low-latency decision-making, and complex workloads—all directly on device.
Compared to the Jetson AGX Orin™, Jetson Thor provides:
- 7.5× more AI compute
- 3.5× better energy efficiency
- 2× the memory
This makes it a perfect fit for agentic AI, reinforcement learning, and advanced robotics workloads.
The NVIDIA Jetson AGX Thor Developer Kit are now generally available to empower developers to build the future of physical AI. The Jetson AGX Thor Developer Kit starts at $3,499 and is available through Seeed Studio.
The Future Is Now: Get Hands-On with Jetson Thor at Embodied AI Hackathons
The global robotics community is coming together to put Jetson Thor into action. The Embodied AI Worldwide Hackathon, co-hosted by Seeed Studio, NVIDIA, and Hugging Face, is giving developers hands-on access to the Jetson Thor platform, along with a full ecosystem of developer tools.
This hackathon will feature on-site events in Shenzhen, China, and the Bay Area, U.S., with a focus on building Home Task Robots and Cooking Robots. Participants will get hands-on experience with:
- Jetson Thor Developer Kit
- NVIDIA Isaac Sim for simulation
- Isaac GR00T N1.5, an open foundation model for generalized humanoid robot reasoning and skills.
- LeRobot by Hugging Face, an open-source robotics library
- SO-ARM 101 Open-source 6-Axis 2 robot arms for end-to-end imitation learning
- FashionStar StarAI Arm: Viola & Violin Pre-assembled 7DoF robot arms for flexible and precise motion
Developers will also be able to experiment with reinforcement learning, imitation learning, and Vision-Language-Action (VLA) models like Isaac GR00T N1.5. The goal: to design affordable, capable robots that can handle everyday challenges.
📍 Event Dates
- October 18–19: Shenzhen, China → Register now
- October 25–26: Bay Area, U.S. → Register now
Building What’s Next
Modern robots demand enormous computing power to process multiple sensor streams and make intelligent decisions in real time. With Jetson Thor, developers, researchers, and enthusiasts can push the boundaries of what’s possible in robotics.
🚀 Now is the time to get involved.
- Explore the Jetson AGX Thor Developer Kit at Seeed Studio.
- Sign up for the upcoming hackathons and join the global community building the future of robotics.
Seeed NVIDIA Jetson Ecosystem
Seeed is an Elite partner for edge AI in the NVIDIA Partner Network. Explore more carrier boards, full system devices, customization services, use cases, and developer tools on Seeed’s NVIDIA Jetson ecosystem page.
About Author
