ASUS IoT leverages NVIDIA Jetson Orin to double GenAI performance for edge applications
ASUS IoT, edge AI computers now support “Supermode” for NVIDIA Jetson Orin NX and Orin Nano series, doubling generative AI (GenAI) model performance.
Supermode enhances AI TOPS by up to 1.7X and memory bandwidth by up to 1.5X, enabling faster processing for large language models (LLMs), visual language models (VLMs), and vision transformers (ViTs).
Jetson Orin modules are compatible with popular machine learning frameworks like Hugging Face Transformers, NVIDIA TensorRT-LLM, and more, making them ideal for edge applications and robotics.
ASUS IoT models (PE1102N, PE1101N, PE1100N) are designed for compactness, versatility, and ruggedness, suitable for smart city, transportation, and logistics applications. These devices offer faster data processing, enhanced security, and reduced total cost of ownership (TCO) for edge AI workloads.
The systems are built for extreme conditions, with wide temperature range operation and excellent thermal performance for outdoor use.
Supermode is enabled via the latest NVIDIA JetPack 6.2 SDK, which includes updates like CUDA 12.6, TensorRT 10.3, and Ubuntu 22.04-based systems. JetPack 6.2 boosts AI inference performance for GenAI models by up to 2X and supports high-power Supermodes for Jetson Orin Nano and NX modules.
ASUS continues to innovate in AIoT solutions, focusing on sustainability and creating a net-zero enterprise for a circular economy.
Article Topics
AI/ML | AIoT | ASUS | edge AI | edge AI servers | generative AI | IoT | Nvidia
Comments