Apple is developing a new Mac app called Harbor that enables users to run local AI models directly on their hardware. According to leaker Aaron Perris, the app transforms Macs into private AI inference servers, allowing open-weight models to run entirely on Apple silicon without data leaving the device or touching the cloud. The app supports connecting multiple Thunderbolt 5-capable Macs to form a unified cluster, pooling unified memory for larger models and higher concurrency.
Key features include instant setup in minutes—Harbor automatically detects devices over Thunderbolt 5 and begins running without manual configuration or command-line expertise. Users can browse, download, and load open-weight models directly from the app, eliminating the need for separate model management. The Harbor app is currently in private preview under Apple's Claris brand and is expected to launch publicly in the coming weeks and months.
Source: 9to5Mac · Summarized by HeadlinesBriefing