Login to Continue Learning
[NVIDIA’s GB10 Superchip is gaining traction as Acer has adopted it for its Veriton GN100 AI Mini PC, starting at $3999.](https://news.acer.com/acer-unveils-the-veriton-gn100-ai-mini-workstation-built-on-the-nvidia-gb10-superchip)
Acer Packs NVIDIA’s GB10 Superchip Inside Its Veriton GN100 AI Mini PC, 20 Cores & 128 GB Memory Priced at $3999
Back in May, NVIDIA launched its GB10-powered DGX Spark systems, which are small AI boxes with significant AI capabilities. At Hot Chips 2025, NVIDIA provided a deep dive into the GB10 Superchip architecture. This low-power solution for AI workloads has led to Mini PCs from OEMs like ASUS, Gigabyte, MSI, and others. Today, Acer joins this lineup with its own GB10-powered Mini PC, the Veriton GN100.
- Ultra-compact personal AI workstation featuring the NVIDIA GB10 Grace Blackwell Superchip, 128 GB of unified memory, 4 TB storage, and up to 1 PFLOPS of FP4 AI performance.
- Designed for local large AI model work and scalable to larger workloads. Connect two Veriton GN100 systems with NVIDIA ConnectX-7 SmartNIC. Pre-installed with the latest NVIDIA AI software stack and support for common developer tools like PyTorch, Jupyter, and Ollama.
- Space-efficient and secure, with Kensington lock support and the ability to link two units for handling up to 405 billion parameter models.
Starting with the specifications, the Acer Veriton GN100 AI Mini PC is based on the NVIDIA GB10 AI Superchip. This SoC includes a 20-core ARM CPU and a Blackwell GPU with 6144 cores, delivering up to 1 PFLOPs of NVFP4 AI performance. The system comes equipped with 128 GB of LPDDR5x memory, shared between the CPU and GPU, plus 4 TB of NVMe storage. It also features a good set of I/O ports: four USB 3.2 Type-C, HDMI 2.1b, LAN port, Wi-Fi 7, Bluetooth 5.1, and an NVIDIA ConnectX-7 Smart NIC.
The NIC is crucial for combining two Veriton GN100 Mini PCs to expand AI capabilities. Each system measures 150 x 150 x 50.5 mm (5.91 x 5.91 x 1.99 inches) and weighs less than 1.5 kg, making it compact yet powerful. Combining two allows running up to 405 billion parameter LLMs with ease.
Pricing starts at $3999 US, EUR 3999, and AUD 6499 for the base configuration. A $4K price tag is high but matches what NVIDIA charges for its Spark system. 1 PFLOPs of AI performance in a compact Mini PC form factor is impressive, given that larger workstations with similar power would require additional components at higher costs.
| Name | Acer Veriton GN100 AI Mini Workstation |
| Model | GN100 |
| Operating System | DGX Base OS & NVIDIA AI Software Stack |
| Processors | 20-core Arm-based processor, up to 1 PFLOPs of FP4 AI performance |
| Graphics | NVIDIA GB10 Grace Blackwell Superchip |
| Memory | 128 GB LPDDR5x Coherent Unified System memory |
| Storage | Up to 4 TB M.2 NVMe with self-encryption |
| Audio | HDMI multi-channel audio output |
| Ports | Four USB 3.2 Type C, HDMI 2.1b, RJ-45 connector, NVIDIA ConnectX-7 Smart NIC |
| Networking | Wi-Fi 7, Bluetooth 5.1 or above |
| Network Interface Card | NVIDIA ConnectX-7 Smart NIC |
| Security | Kensington lock, Local AI Model Execution (data privacy) |
| Dimensions/Weight | 150 (W) x 150 (D) x 50.5 (H) mm (5.91 x 5.91 x 1.99 inches)/ less than 1.5 kg |
