Advantech Expands Edge AI Portfolio With New NVIDIA Jetson Chips for Robots
Quick Reads
- Advantech adds NVIDIA Jetson T2000 and T3000 to its edge AI portfolio
- Joins existing T4000 and T5000 platforms for a full range of options
- T3000 hits 865 FP4 TFLOPS at 70W; T2000 offers 400 FP4 TFLOPS at 40W
- Commercial availability expected Q1 2027
- Targets robotics, autonomous machines, and industrial automation
Advantech just made its edge AI portfolio a lot more flexible. The company is adding NVIDIA Jetson T2000 and T3000 modules to its lineup, giving businesses more ways to bring serious AI computing power directly to the robots and machines that need it, without leaning on the cloud.
Factories and warehouses increasingly want machines that can see, think, and react on their own. That takes computing muscle packed into a small, efficient package, exactly the gap this edge AI portfolio update fills. Advantech’s new modules provide customers with a complete range of AI computing platforms from entry level commercial systems to flagship industrial solutions.
The T3000 Is The Standout
It delivers up to 865 FP4 TFLOPS of AI performance with 32GB LPDDR5X memory at just 70W, offering similar inference performance to the existing Jetson T5000 while cutting power consumption and memory costs. The T2000, meanwhile, is the budget-friendly option: it offers 400 FP4 TFLOPS with 16GB LPDDR5X memory within a 40W power envelope, serving as the entry point to the Thor family for cost-effective deployment at scale.
Together with the existing T4000 and T5000, Advantech now covers the full Jetson Thor spectrum, letting customers pick exactly what fits their budget and performance needs. The hardware spans MIC-AI industrial systems, MIB integration boards, AFE and ASR robotics controllers, and the AIR vision series for factories and smart cities.
Commercial availability is expected in the first quarter of 2027. It’s a preview of what’s coming, but one that could already shape robotics and automation roadmaps being planned today.
