Login to Continue Learning
NVIDIA has released its latest Jetson system, the Jetson Thor, which is powered by the Blackwell architecture. This new platform is designed to advance the era of humanoid robots, enhancing their capabilities and enabling them to perform more complex tasks independently.
NVIDIA’s Newest Jetson Thor Platform Brings Massive Upgrades in Compute Capabilities, But at a High Cost
NVIDIA CEO Jensen Huang has identified ‘Physical AI’ as a significant focus for the company. This initiative not only aims to boost revenue but also to enable robots to reason and act more autonomously through advanced AI techniques. According to a new blog post, NVIDIA has launched the Jetson Thor modules, which offer significantly higher compute capabilities compared to the previous Jetson Orin module. These upgrades are particularly important for advancing humanoid robotics.
Technical Details of the Jetson Thor
The Jetson Thor is equipped with the new Blackwell-based Jetson T5000 module, featuring 14 ARM Neoverse-V3AE CPU cores and a 2560-core GPU. NVIDIA claims that this platform provides 7.5x more AI compute power, 3.1x more CPU performance, and twice as much memory compared to the previous generation. This capability allows it to run advanced generative AI frameworks and reasoning models.
Key Specifications of Jetson Thor
– **Module:** Jetson T5000
– **GPU:** 2560-core NVIDIA Blackwell architecture GPU with 96 fifth-gen Tensor Cores
– **CPU:** 14-core Arm Neoverse V3AE 64-bit CPU
– **Memory:** 128 GB 256-bit LPDDR5X, 273 GB/s
– **Networking:** 1x 5GbE RJ45 connector, 1x QSFP28 (4x 25 GbE)
– **I/O:** QSFP connector, HDMI port, DisplayPort, 2x USB-A 3.2, 2x USB-C 3.1, Gigabit Ethernet, 2x 13-pin CAN header, Microfit power jack
– **Storage:** 1 TB NVMe M.2 Key M slot
– **Power:** 40 W – 130 W
Adoption and Cost Considerations
Agility Robotics and Boston Dynamics are among the leading robotics companies that have started using Jetson Thor systems in their products, indicating potential future advancements in capable humanoid robots.
However, the high cost of these systems poses a significant challenge. NVIDIA has priced the Jetson AGX Thor developer kit at $3,499, which is expensive when considering the overall costs involved in building humanoid robots.
Commitment and Future Plans
Jensen Huang personally signed many initial units of the Jetson Thor sent to customers, demonstrating his commitment to this segment of the AI industry. Additionally, NVIDIA’s key partner Foxconn plans to introduce humanoid robots into the market by year-end through a partnership, highlighting the potential for physical AI to become the next big trend in the tech industry.
This advancement underscores the growing importance of Physical AI and its potential to transform various sectors through advanced robotics technology.
