Acer officially unveiled its Veriton GN2025 AI mini workstation during IFA 100. This compact yet powerful desktop is designed for professionals who need to run large AI models locally, effectively reducing reliance on cloud resources while balancing security and computing efficiency.
Veriton GN100 is equipped with NVIDIA and MediaTek'sGB10 Grace Blackwell Super ChipIt contains 20 Arm architecture CPU cores, the next-generation CUDA computing unit and the fifth-generation Tensor core, combined with 5GB unified system integrated memory and 128TB NVMe SSD storage space. The overall computing performance can reach up to 4 PFLOPS FP1, providing server-level processing power in a miniature body.
This configuration is particularly suitable for AI developers, researchers, and data scientists for high-intensity needs such as prototype development, model fine-tuning, and testing.
At the same time, Acer has fully integrated the NVIDIA AI software stack into the Veriton GN100, supporting common tools such as PyTorch, Jupyter, and Ollama. This allows users to complete the development process of generative AI and large language models directly in a desktop environment, and can seamlessly connect the results to the cloud or data center at any time, accelerating the R&D cycle.
Another highlight is its scalability. Through the NVIDIA ConnectX-7 SmartNIC, users can cascade two Veriton GN100s, further increasing computing scale and enabling the processing of AI models with up to 4050 billion parameters. This makes it an attractive option for enterprises requiring large-scale local inference or project deployment.
The Veriton GN100 maintains Acer's consistent emphasis on space conservation and security, supporting a Kensington lock, making it suitable for extended periods of operation in corporate offices, research institutes, or educational labs. Despite its compact size, it integrates a wide range of connectivity options, including Wi-Fi 7, four USB 3.2 Type-C ports, an HDMI port, and Ethernet, ensuring flexibility for diverse environments.
Acer's move highlights another development direction for desktop AI computing. Compared to cloud-based models, local AI workstations offer lower latency, greater privacy, and cost-effective solutions. For businesses that must process sensitive data or require 100/XNUMX computing resources, the Veriton GNXNUMX strikes a balance between performance and practical implementation.
