Developers to Gain Access to Apple’s On-Device AI Models

Apple is set to expand its artificial intelligence (AI) ecosystem by granting third-party developers access to its on-device AI models, a move aimed at enhancing app functionalities across its platforms, Bloomberg is reporting.

This initiative is expected to be officially announced at the upcoming Worldwide Developers Conference (WWDC) on June 9, 2025.

The company plans to release a Software Development Kit (SDK) that will allow developers to integrate Apple’s smaller, on-device large language models (LLMs) into their applications. These models power features within Apple’s “Apple Intelligence” suite, including tools for text summarization, rewriting, and image generation.

Currently, developers have limited access to Apple Intelligence features, such as AI Writing Tools and the Image Playground generator. The forthcoming SDK will provide deeper integration capabilities, allowing for more customized and responsive app experiences.

Apple’s approach to AI emphasizes tight integration with its hardware and a focus on user privacy. The company has previously released Core ML models on platforms like Hugging Face, enabling developers to create applications with advanced image and text processing capabilities. These models are optimized for performance and designed to operate entirely on users’ devices.

The anticipated announcement at WWDC is also expected to coincide with a major update, codenamed “Solarium,” aimed at unifying iOS, macOS, and iPadOS to mirror the Vision Pro operating system.

As Apple continues to invest in AI technologies, the opening of its on-device models to developers marks a significant step towards a more innovative and privacy-conscious app development environment.