Acer launches the Veriton GN100 AI Mini workstation, a compact system with Nvidia GB10 Superchip for local processing of large AI models and optional scalability through coupling of two devices.
During IFA 2025, Acer introduced the Veriton GN100 AI Mini workstation, a compact PC with server capabilities thanks to the new Nvidia GB10 Grace Blackwell superchip. The system is designed for local AI applications and supports scalability through coupling with a second device. The Veriton GN100 AI Mini Workstation is available for 3,999 euros including VAT.
AI Performance in Mini Format
The Veriton GN100 is aimed at developers and researchers who want to run large AI models locally. The device contains the Nvidia GB10 superchip with 20 Arm-based CPU cores, CUDA cores, and the latest generation Tensor cores. The workstation features 128 GB of unified memory and 4 TB NVMe SSD storage. According to Acer, the device delivers up to 1 PFLOPS of FP4 AI performance.
Thanks to this configuration, users can perform AI tasks locally without relying on cloud infrastructure. This not only reduces costs but also offers advantages in terms of data sovereignty and latency. The system comes with the full Nvidia AI software stack, including support for commonly used tools such as PyTorch, Jupyter, and Ollama.
Expandable
For heavier AI workloads, two GN100 units can be linked together via the Nvidia ConnectX-7 SmartNIC. This creates a system capable of handling AI models up to 405 billion parameters. This scalability makes the device suitable for applications such as training and running large language models.
The device has built-in security options such as a Kensington lock. In terms of connectivity, it includes Wi-Fi 7, four USB 3.2 Type-C ports, HDMI, and Ethernet.