ASUS today unveiled the UGen300 USB AI Accelerator, its first dedicated USB device for AI tasks. This compact module, measuring just 105 x 50 x 18 mm, houses the Hailo-10H AI processor delivering 40 AI TOPS of dedicated performance. It features 8 GB LPDDR4 memory and connects via USB-C, consuming only 2.5 watts. The UGen300 enables cross-platform AI inference on Windows, Linux, and Android devices without cloud dependence, supporting frameworks like TensorFlow and PyTorch.

By placing powerful AI acceleration directly on USB peripherals, ASUS addresses the growing demand for on-device generative AI. Users gain unmatched edge performance, running demanding applications like text generation and video summarization locally. This avoids cloud latency, subscription costs, and privacy concerns associated with remote processing. The plug-and-play design makes it ideal for developers, educators, and embedded-system builders needing scalable, portable AI power.

The UGen300 empowers systems lacking native AI capabilities to handle complex tasks efficiently, freeing host resources. Its dedicated RAM architecture prevents bottlenecks, ensuring consistent throughput for demanding AI pipelines. ASUS positions this as a practical solution for bringing advanced AI capabilities to everyday devices and edge computing scenarios.