TECH – Acer has introduced the Veriton GN100 AI Mini Workstation, a remarkably compact yet high-performance device tailored to handle demanding AI workloads in a desktop format. At just 150 × 150 × 50.5 mm, the GN100 houses Nvidia’s ultra-powerful GB10 Grace Blackwell Superchip, making it akin to a supercomputer in miniaturized form.
Powered by a 20 -core ARM processor—with 10 Cortex-X925 and 10 Cortex-A725 cores—and equipped with a Blackwell-based GPU, the unit delivers up to 1 petaFLOP of FP4 AI performance. It features a substantial 128 GB of unified LPDDR5x memory and offers up to 4 TB of self-encrypting M.2 NVMe storage, spread across four slots.
Connectivity is comprehensive: four USB 3.2 Type-C ports, an HDMI 2.1b port, an RJ-45 Ethernet jack, Wi-Fi 7, and Bluetooth 5.1. Notably, it incorporates a Nvidia ConnectX-7 SmartNIC, enabling two GN100 units to interconnect and jointly handle AI models with up to 405 billion parameters—a remarkable capability for such a small form factor.
Read More: Vikram Chip Debut Marks India’s Semiconductor Breakthrough
On the software side, the Veriton GN100 supports Nvidia’s AI software stack, including CUDA, cuDNN, and TensorRT, and it’s compatible with popular frameworks such as TensorFlow, PyTorch, MXNet, and JAX—making it suitable for developers, researchers, and students.
High-performance benchmarks aside, Acer also emphasizes practicality and data security: the device includes a Kensington lock slot and ensures local AI execution, reducing reliance on cloud services and minimizing latency risks.
Priced at US$3,999, the Veriton GN100 sits alongside comparable products from other major OEMs like Asus, Dell, HP, and MSI, who also offer variants based on Nvidia’s Project Digits superchip infrastructure.
With its blend of compact design, top-tier AI compute strength, and robust software support, the Acer Veriton GN100 delivers extraordinary computing capability in a small footprint redefining what a “mini PC” can achieve in AI environments.
