Alibaba’s update of Qwen3 models likely heralds Apple Intelligence coming to China: report

Alibaba’s (NYSE:BABA) update of its Qwen3 open-source, artificial intelligence models for Apple’s (NASDAQ:AAPL) MLX architecture, likely signals the eventual deployment of Apple Intelligence features to mainland China, according to the South China Morning Post.

“Excited to launch Qwen3 models in MLX format today!,” Alibaba’s Qwen group posted on X on Monday. “Now available in 4 quantization levels: 4bit, 6bit, 8bit, and BF16 — Optimized for MLX framework.”

With the update, the Qwen3 models can run on Apple’s iPhone, iPad, MacBook and Mac. They were released yesterday on Hugging Face.

“This latest development appears to reflect Apple’s efforts to expand the availability of its Apple Intelligence suite to mainland China,” the report said. “Under Chinese rules, any models that are part of Apple Intelligence would need regulatory approval for public release in the country.”

Alibaba owns the South China Morning Post.

Due to China’s regulations, Apple has partnered with Alibaba to deliver Apple Intelligence features to customers in China. Everywhere else, Apple Intelligence relies on OpenAI’s GPT models. Apple Intelligence refers to an array of AI tools, such as Genmoji and Writing Tools, for users of Apple hardware.

Seeking Alpha reached out to Apple for comment.

According to some reports, Apple’s launch of its Apple Intelligence features in China is being held up by the trade war between China and the U.S.

Apple’s iPhone sales took the top spot in China in May, with global sales rising 15% year-over-year during April and May, according to data from Counterpoint Research. China ranks as Apple’s largest market outside the U.S.