ASUS IoT has announced its Edge AI systems now support the new Super Mode under the NVIDIA Jetson Orin platform.
This software update requires no hardware modification and can boost AI TOPS performance by up to 1.7x, memory bandwidth by up to 1.5X, and performance of generative AI models by as much as 2X which means things like LLMs, VLMs, and ViTs can run much faster and more efficient.
Jetson Orins are also known for their massive compatibility with popular machine learning frameworks, including Hugging Face Transformers, Ollama, llama.cpp, vLLM, MLC, and NVIDIA TensorRT-LLM, making it super flexible for all sorts of edge applications, robotics, and multimodal systems.
The recipients for this new Super Mode update include the PE1102N, PE1101N, and PE1100N which are widely used for things like smart cities, smart transportation, and logistics applications such as smart parking, traffic analysis, and automated guided vehicles (AGVs) or autonomous mobile robots (AMRs), citing great TCO figures, beefy security, and tanky build quality as their key selling point.
The latest NVIDIA JetPack 6.2 SDK supports Super mode on Jetson Orin NX and Orin Nano series modules, which can be activated on ASUS IoT edge AI computers. JetPack 6.2 introduces production-level features, including Jetson Linux 36.4.3, CUDA 12.6, NVIDIA TensorRT 10.3, and other AI tools, boosting AI inference performance for generative AI models by up to 2X.
Availability
For Singapore, the ASUS IoT PE1102N is currently available for inquiries and orders.
