Acer has announced that its much-anticipated Veriton GN100 AI mini workstation will be available this December. Starting at $3,999, the compact system delivers server-class power in a desktop-sized form factor, making it a serious option for developers, researchers, and AI enthusiasts.
At the heart of the GN100 is NVIDIA’s GB10 Grace Blackwell Superchip, providing up to 1 PFLOPS of FP4 AI performance. The system pairs a 20-core Arm-based CPU with next-generation CUDA and fifth-generation Tensor Cores. Backed by 128GB of unified LPDDR5x memory and up to 4TB of NVMe M.2 SSD storage, it’s designed to handle everything from large language model fine-tuning to complex data analysis.
Acer includes the NVIDIA AI software stack and DGX Base OS, giving users immediate access to common frameworks like PyTorch and Jupyter, as well as local LLM environments such as Ollama. This allows developers to prototype, train, and deploy AI models without relying on the cloud. Obviously, this is a huge win for those prioritizing data privacy and cost control.
The Veriton GN100 can also scale up. Thanks to the integrated NVIDIA ConnectX-7 Smart NIC, two units can be linked together to handle AI models reaching up to 405 billion parameters. Connectivity options include Wi-Fi 7, four USB 3.2 Type-C ports, HDMI 2.1b, and an Ethernet jack, while a Kensington lock ensures physical security.
Despite its power, the GN100 weighs less than 1.5 kilograms and measures just 150mm square. Acer says the system is ideal for professionals who need portable AI horsepower without maintaining a full data center.
2 thoughts on “Acer announces availability of Veriton GN100 AI mini workstation powered by NVIDIA Grace Blackwell Superchip”