NVIDIA has unveiled its brand-new Jetson Thor solutions, expanding into mainstream robotics, humanoids & Edge AI with the T3000 & T2000.
NVIDIA Jetson Thor T3000 Offers 865 TFLOPs of AI Compute For Humanoids & Robotics, While T2000 Pushes 400 TFLOPs For Edge AI
We already told you that some big news was coming on the Physical AI side of things when Jensen landed in Japan yesterday, and the first of these announcements has been revealed in the form of the NVIDIA Jetson Thor T3000 and T2000.
Kawasaki Heavy Industries provides technology designed to improve the overall efficiency of hospital operations, including with its FORRO, Nyokkey and NURABOT robots.
The company plans to use NVIDIA Holoscan IGX, Isaac for Healthcare, Isaac GR00T and Cosmos to develop surgical support functions, nursing assistant and hospital transport robots.
Direava is developing a surgical vision language model for real-time surgical video understanding and natural language interaction with surgical scenes. Direava aims to evolve this technology into an intelligence layer for future surgical AI and physical AI in the operating room.
Building upon the NVIDIA Jetson AGX Thor platform, the T3000 and T2000 computers drive mass-market robotics and edge AI applications at scale. To name a few, X, Agility, Agile Robots, Amazon Robotics, Boston Dynamics, FANUC, Hitachi, Medtronic, and Techman Robo, are some of the tech firms that are using these solutions to advance the Physical AI spectrum.
Jetson Thor T3000 - Primed at Humanoids & Robotics
The NVIDIA Jetson Thor T3000 is a streamlined version of the flagship Thor T5000 solution. It packs 865 TFLOPs of FP4 (AI) compute in a form factor that is roughly half the size and power of the T5000. Just like the higher-end solutions, the Thor T3000 packs an NVIDIA Blackwell GPU, up to eight Arm Neoverse CPU cores, 32 GB of LPDDR5X memory with 273 GB/s of bandwidth, 25 GbE connectivity, and the same safety features when running the NVIDIA Halos Robotics stack.
NVIDIA claims that the T3000 offers roughly the same inference performance as the T5000 in multimodal workloads such as LLMs, VLMs, and World Foundation Models. The T3000 solution helps avert the rising memory costs with a balanced platform. The T3000 platform is rated at 70 Watts.
Jetson Thor T2000 - Made For Edge AI
For Thor T2000, NVIDIA is packing 400 TFLOPs of FP4 compute with 16 GB of memory. The T2000 platform aims to be an entry-level solution for Visual AI agents, autonomous mobile robots, and various intelligence-based operations. The T2000 platform is rated at 40 Watts.
These new add-ons to the Jetston AGX Thor family now round up the product roadmap, covering all segments from the entry-level Orin computers to the higher-end Jetston Thor platforms.
NVIDIA Unlocks Further Optimizations for Jetson Devices, Reducing Memory Usage Significantly
NVIDIA has also released its new Jetson agent skills, which developers can leverage to optimize the entire software stack, resulting in significant memory savings. This allows customers to move one SKU down, saving costs and power.
NVIDIA has shown various examples where Jetson AGX Orin achieves the same results within Industrial and Humanoid use-cases while reducing memory usage by 50%.
Humanoid robotics leaders including UBTech and Agile Robots, along with industrial solutions provider Connect Tech, have reduced memory usage by up to 15GB, enabling them to move from NVIDIA Jetson AGX Orin 64GB to the 32GB module.
In smart retail, SandStar reduced memory usage by up to 4GB, enabling deployment on the NVIDIA Jetson Orin NX 8GB module instead of the 16GB configuration. In companion robotics, GROOVE X, creator of the LOVOT robot, uses Jetson’s heterogeneous AI accelerators to optimize workload distribution, reducing memory usage and enabling deployment on lower-memory configurations.
In intelligent transportation, NoTraffic reduced memory usage by 30% on Jetson TX2 NX, creating headroom to add more AI capabilities into its smart traffic platform without increasing hardware requirements.
The Jetson Thor T3000 will be available for use in emulation mode this month with JetPack 7.2.1, while T2000 will be available in a future release. Both modules are scheduled for launch in the first quarter of 2027. Partners providing these Thor-based solutions include ADLINK, Advantech, AAEON, Aetina, Auvidea, AVerMedia, Connect Tech, ForeCR, JWIPC, NEXCOM Robotic Solutions, Realtimes, Seeed Studio, Twowin, TZTEK, and YUAN.
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
