Acer's New Workstation Brings a Frankly Ridiculous 1 Petaflop AI Performance to Your Desktop - Get an Nvidia Grace Blackwell in Your Business Device Now

Why It Matters

By delivering server‑grade AI compute in a compact desktop, the GN100 enables enterprises to keep sensitive data in‑house, cut reliance on costly cloud services, and accelerate AI prototyping with predictable expenses. This could shift a segment of AI workloads from the cloud to the edge, reshaping enterprise AI deployment strategies.

Summary

Acer unveiled the Veriton GN100 AI mini‑workstation, a 1.2 kg desktop that packs Nvidia’s GB10 Grace Blackwell Superchip and claims up to 1 petaflop of FP4 AI performance. The unit offers up to 128 GB of LPDDR5x unified memory, 4 TB of self‑encrypting NVMe storage, and can be paired with another GN100 via Nvidia ConnectX‑7 to run models as large as 405 billion parameters. Running NVIDIA DGX OS with support for PyTorch, Jupyter and Ollama, it targets edge AI workloads for UK enterprises seeking on‑premise inference, reduced latency and data‑privacy. Pricing starts at €3,999, with early‑access registration open for UK organisations.