Apple Intelligence Foundation Language Models: Tech Report 2025

TL;DR

This paper introduces two advanced multilingual multimodal foundation language models by Apple, optimized for on-device and server deployment, demonstrating high performance, privacy, and developer-friendly features across Apple devices and services.

Contribution

The paper presents novel architectures for on-device and server models, large-scale training methods, and a new framework for easy integration and fine-tuning of foundation models within Apple's ecosystem.

Findings

Models match or surpass open baselines in benchmarks.

Supports additional languages and multimodal understanding.

Ensures privacy and responsible AI practices.

Abstract

We introduce two multilingual, multimodal foundation language models that power Apple Intelligence features across Apple devices and services: i a 3B-parameter on-device model optimized for Apple silicon through architectural innovations such as KV-cache sharing and 2-bit quantization-aware training; and ii a scalable server model built on a novel Parallel-Track Mixture-of-Experts PT-MoE transformer that combines track parallelism, mixture-of-experts sparse computation, and interleaved global-local attention to deliver high quality with competitive cost on Apple's Private Cloud Compute platform. Both models are trained on large-scale multilingual and multimodal datasets sourced via responsible web crawling, licensed corpora, and high-quality synthetic data, then further refined with supervised fine-tuning and reinforcement learning on a new asynchronous platform. The resulting models support several additional languages while understanding images and executing tool calls. In public benchmarks and human evaluations, both the server model and the on-device model match or surpass comparably sized open baselines. A new Swift-centric Foundation Models framework exposes guided generation, constrained tool calling, and LoRA adapter fine-tuning, allowing developers to integrate these capabilities with a few lines of code. The latest advancements in Apple Intelligence models are grounded in our Responsible AI approach with safeguards like content filtering and locale-specific evaluation, as well as our commitment to protecting our users' privacy with innovations like Private Cloud Compute.