According to CoinMeta, Apple has launched a new Mac, positioned as a low-cost AI machine capable of running large models locally, challenging the cloud computing solutions of Nvidia and Microsoft. Apple has begun shipping upgraded versions of Mac, mini, and Mac Studio, aiming to provide a more cost-effective alternative for AI inference workloads. According to Reuters, Apple demonstrated four Mac Studio running a trillion-parameter AI model under a single wall socket. This strategy brings Apple more directly into cost competition with Nvidia and Microsoft. Apple believes that having powerful hardware can change the cost structure of running workloads locally, and its unified memory architecture allows CPU and GPU to share memory, simplifying the operation of large AI models. Apple is also developing enterprise AI servers, which may extend this efficiency argument to data centers. Reuters cites IDC data indicating that Apple's market share in the enterprise desktop market is only 4.6%, while Windows systems account for about 91%. Nevertheless, as enterprises question whether AI tasks should be placed in expensive cloud environments, Apple's economics could become interesting.