Introducing Apple’s On-Device and Server Foundation Models
In this article:
- Apple Intelligence consists of multiple highly-capable generative models tailored for users’ everyday tasks and adaptable to their current activities.
- These foundation models are fine-tuned for various user experiences, such as writing and refining text, prioritizing and summarizing notifications, creating playful images for conversations, and simplifying interactions across apps.
- This overview will detail the development and adaptation of two specific models:
- A ~3 billion parameter on-device language model.
- A larger server-based language model available with Private Cloud Compute, running on Apple silicon servers.
- Both models are designed to perform specialized tasks efficiently, accurately, and responsibly.
