Acer’s New SFF RTX Spark Desktop Targets Local AI Workloads
AI-summarised brief · reviewed before publication
Acer unveiled the SFF RTX Spark, a compact desktop built around NVIDIA’s new RTX Spark platform, at its next@Acer showcase during IFA 2026. The machine integrates NVIDIA’s RTX Spark superchip, which pairs a Blackwell RTX GPU with up to 6,144 CUDA cores and a 20‑core Grace CPU, linked via NVLink‑C2C and equipped with fifth‑generation Tensor Cores for FP4 AI tasks. NVIDIA rates the configuration at up to one petaflop of FP4 performance and claims it can run AI models of up to 120 billion parameters locally. The system supports up to 128 GB of unified LPDDR5X memory, allowing shared access for CPU and GPU, and includes the full RTX software stack—CUDA, TensorRT, OptiX, DLSS 5, Reflex 2 and G‑SYNC—targeting AI developers, creators, and gamers. Pricing and release timing remain undisclosed.
💡 Why It Matters
- · By delivering petaflop‑class AI compute in a desk‑friendly chassis, the SFF RTX Spark lets users run large models without relying on cloud services, reshaping how creators and developers access high‑end AI power.
