OpenAI and rival AI labs are buying tens of thousands of Mac minis to train computer-use agents

According to The Information, OpenAI has purchased tens of thousands of Mac minis and Mac Studios to train computer agents. Anthropic also relies on Apple hardware. Demand is so high that the most powerful models have been sold out for months. Apple’s Mac revenue rose by nearly 29 percent to $10.4 billion in the June quarter. The article OpenAI and rival AI

The top 3

  1. Google's TPU Dominance in Cloud AI: Google's Tensor Processing Units (TPUs) are custom-designed ASICs that offer high performance and efficiency for machine learning workloads, particularly within Google Cloud, powering services like Google Search and AlphaGo. They are optimized for matrix operations, crucial for neural networks, and provide scalability and energy efficiency.
  2. FPGAs for Flexible AI Acceleration: Field-Programmable Gate Arrays (FPGAs) provide reconfigurable hardware for AI acceleration, offering a balance between the flexibility of CPUs and the performance of ASICs, suitable for evolving AI algorithms and real-time edge AI applications.
  3. Custom ASICs for Niche AI Tasks: Many tech giants and startups, including OpenAI with its "Jalapeño" chip, are developing custom Application-Specific Integrated Circuits (ASICs) optimized for particular AI tasks, especially large language model inference, to achieve maximum efficiency and reduce reliance on general-purpose GPUs.

Sources

Open the full topic