Apple has trained a large language model for China with support from Alibaba, according to three people familiar with the matter.

The move marks a shift from Apple’s plan to rely mainly on Chinese partners’ models for AI features on iPhones and other devices sold there.

The Cyberspace Administration of China, the country’s internet regulator, registered Apple’s generative AI service in July 2026, clearing the way for Apple Intelligence to launch on Chinese iPhones in the coming months.

In China, Apple is also expected to use Alibaba’s Qwen model and technology from Baidu.

It remains unclear how Apple’s self-trained model will work alongside those Chinese models.

The arrangement reflects China’s rules for public generative AI services and helps explain why Apple cannot use OpenAI or Anthropic in the country.

Apple has said Apple Intelligence keeps as much processing as possible on device and sends more complex requests to Private Cloud Compute.

The China-specific model could give Apple more control over the AI experience in a market where it faces strong competition from Huawei and other local smartphone brands.

The lack of AI features on Chinese iPhones had been seen as a factor weighing on Apple’s sales there.

🔗 Source: Reuters

Recent Apple developments