AI Labs Fuel Memory Shortage as OpenAI Buys Tens of Thousands of Mac Minis and Nvidia's RTX Spark Sells Out Pre-Launch

OpenAI has purchased tens of thousands of Apple Mac minis and Mac Studios to support its AI training efforts. This massive order is separate from OpenAI's usual cloud GPU cluster rentals, indicating a preference for these compact desktops for specific tasks.

The purchases are part of the growing demand for memory-rich hardware in AI labs. Unlike gaming PCs, which focus on raw processing power, AI developers prioritize memory to run large language models locally. Apple's unified memory architecture allows Mac minis and Mac Studios to hold larger model sizes than typical discrete graphics cards.

Nvidia's upcoming RTX Spark platform, a compact AI-PC chip designed for AI workloads, has also seen high demand. Asus, one of the launch partners, confirmed that its initial allocation was pre-sold to distributors months before the hardware reaches store shelves. MSI and other major PC makers are reportedly pushing Nvidia for additional supply.

The RTX Spark's focus on unified memory, with a top-out capacity of 128 GB, makes it an attractive option for AI developers working with large language models. However, this has led to concerns that the hardware may be in short supply before its launch in fall 2026, potentially leaving gamers and PC builders struggling to access these components.

As a result, component supply chains are being reshaped around AI demand, leading to shortages for ordinary consumers. The RTX Spark's steep price tag of around $3,400 may further exacerbate this issue.