Alibaba Group Holding has released its Qwen3 family of artificial intelligence models optimized for Apple’s MLX framework, making them compatible with iPhones, iPads, MacBooks, and other Apple devices.
The announcement was made Monday through a post on WeChat and confirmed on X by the Qwen team under Alibaba Cloud.
The MLX framework, designed for Apple’s silicon hardware, is intended to help developers easily train and run AI models across Apple’s product line. Alibaba’s Qwen3 models are open source, allowing developers to access and modify the source code.
This move may signal preparations for a broader rollout of Apple’s AI features in China. The South China Morning Post, owned by Alibaba, earlier reported that Apple had reached a deal with Alibaba to use its Qwen models in the country. Apple Intelligence, Apple’s AI suite, uses OpenAI’s GPT models outside China, but any AI models used in China must comply with local regulations and gain approval before public release.
