NVIDIA launches the Jetson AGX Thor developer kit.
NVIDIA has just launched the NVIDIA Jetson AGX Thor developer kit and its production modules, which are innovative AI supercomputers aimed at powering millions of robots across diverse sectors, such as manufacturing, logistics, transportation, healthcare, agriculture, and retail.
Early adopters of this technology include prominent companies like Agility Robotics, Amazon Robotics, Boston Dynamics, Caterpillar, Figure, Hexagon, Medtronic, and Meta. Additionally, industry giants such as John Deere, OpenAI, and Physical Intelligence are exploring the Jetson Thor to enhance their research in physical AI.
Jensen Huang, CEO of NVIDIA, highlighted that Jetson Thor is engineered explicitly for robotic system developers, boasting remarkable performance and energy efficiency. This advanced platform can handle multiple generative AI models at the edge, positioning it as a top-tier supercomputer for propelling advancements in physical AI and general robotics. With a powerful GPU architecture and optimized AI frameworks, Jetson Thor is set to redefine capabilities in robotic applications.
The Ultimate Platform for Next-Generation Robotics
The NVIDIA Jetson Thor represents a significant leap compared to its predecessor, the Jetson AGX Orin, boasting an impressive increase of up to 7.5 times in AI compute power and enhancing energy efficiency by 3.5 times. This advanced system-on-module is capable of running a wide array of generative AI models, including cutting-edge vision-language action models like NVIDIA Isaac GR00T N1.5 and various large language models.
One of the standout features of Jetson Thor is its ability to tackle one of the most pressing challenges in robotics: the execution of multi-AI workflows. This capability is essential for enabling robots to engage in real-time, intelligent interactions with humans and their surroundings. By unlocking real-time inference, Jetson Thor is poised to drive innovation across multiple fields, including humanoid robotics, agriculture, and surgical assistance.
This technology has significant implications, as it leads to the development of more intelligent and efficient robotic systems. These systems can function autonomously and interactively within complex environments, ultimately improving productivity and safety across different sectors.
Global Robotics Leaders Build on Jetson Thor
Jetson Thor is powered by the full-stack NVIDIA Jetson software platform, built for physical AI and humanoid robotics, which supports any popular AI framework and generative AI model. It is also fully compatible with NVIDIA’s software stack from cloud to edge, including NVIDIA Isaac for robotics simulation and development, Isaac GR00T humanoid robot foundation models, NVIDIA Metropolis for vision AI, and NVIDIA Holoscan for real-time sensor processing.
Since its launch in 2014, the NVIDIA Jetson platform has truly revolutionized the world of edge AI, attracting over 2 million developers and fostering a vibrant ecosystem with more than 150 partners in hardware systems, software, and sensors. With Jetson Orin, over 7,000 customers across various industries are now harnessing the power of edge AI to drive innovation and efficiency in their operations.
| NVIDIA Jetson AGX Thor Developer Kit Specifications | |
|---|---|
| Technical Specifications: Jetson T5000 Module | |
| AI Performance | 2070 TFLOPS (FP4—sparse) |
| GPU | 2560-core NVIDIA Blackwell architecture GPU with 96 fifth-gen Tensor Cores |
| CPU | 14-core Arm® Neoverse®-V3AE 64-bit CPU 64 KB I-Cache, 64 KB D-Cache 1 MB L2 cache per core 16 MB shared system L3 cache |
| Memory | 128 GB 256-bit LPDDR5X 273 GB/s |
| Storage | 1 TB NVMe M.2 Key M slot |
| Video Encode | 6x 4Kp60 (H.265) 12x 4Kp30 (H.265) 24x 1080p60 (H.265) 50x 1080p30 (H.265) 48x 1080p30 (H.264) 6x 4Kp60 (H.264) |
| Video Decode | 4x 8Kp30 (H.265) 10x 4Kp60 (H.265) 22x 4Kp30 (H.265) 46x 1080p60 (H.265) 92x 1080p30 (H.265) 82x 1080p30 (H.264) 4x 4Kp60 (H.264) |
| Power | 40 W–130 W |
| Reference Carrier Board | |
| Camera | HSB camera via QSFP slot USB camera |
| PCIe | M.2 Key M slot with x4 PCIe Gen5 (populated with 1 TB NVMe) |
| USB | 2x USB-A 3.2 2x USB-C 3.1 |
| Networking | 1x 5 GBe RJ45 connector 1x QSFP28 (4x 25 GbE) |
| Display | 1x HDMI 2.0b 1x DisplayPort 1.4a |
| Other I/O | 2x 13-pin CAN header 2x 6-pin automation header 2x 5-pin JTAG connector 1x 4-pin fan connector—12 V, PWM, and Tach 2x 5-pin audio panel header 2-pin RTC backup battery connector Microfit power jack Power, force recovery, and reset buttons |
| Mechanical | 243.19 mm x 112.40 mm x 56.88 mm (Height includes feet, carrier board, module, and thermal solution) |
The latest advancement, Jetson Thor, takes visual AI to the next level, enabling the development of complex robotic systems, including humanoids and surgical robots. This technology is crucial as it empowers industries to create smarter, more capable robots that can perform intricate tasks and make real-time decisions, ultimately enhancing productivity and safety in sectors ranging from healthcare to manufacturing. The potential applications of this technology are vast, paving the way for a future where intelligent systems seamlessly integrate into our daily lives.
Brett Adcock, CEO of Figure, emphasizes that the advancement of humanoid robots depends on the ability to implement powerful AI models on the robots themselves. He highlights that NVIDIA Jetson Thor’s efficient and compact design enables the deployment of large-scale generative AI models, allowing the robots to perceive, reason, and operate in complex environments effectively.
According to Tye Brady, chief technologist at Amazon Robotics, the future of logistics robotics relies on developing smarter, autonomous systems. He highlights NVIDIA Jetson Thor as essential for providing the necessary computational power and energy efficiency to create AI-powered robots capable of safely navigating dynamic environments, ultimately transforming global goods management.
Availability
The NVIDIA Jetson AGX Thor developer kit starts at $3,499 and is available now. Production modules can be obtained from worldwide distribution partners, while production systems and carrier boards can be purchased from embedded partners. The autonomous vehicle-oriented Thor Drive unit will be available at a later date, featuring the same T5000 and T4000 modules.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
