FuriaHubApple trained its own China LLM with Alibaba’s help, sources tell Reuters
Three people familiar with the matter said Apple built a China-specific model instead of relying only on local third-party systems. Apple and Alibaba did not comment.
On August 14 Reuters reported, citing three people familiar with the matter, that Apple has trained a large language model specifically for China, with support from Alibaba. That is a break from Apple’s earlier plan to power Apple Intelligence in China mainly with domestic partners’ models. Apple and Alibaba declined to comment.
What was already public
US models such as ChatGPT and Claude are not available to pair with Apple Intelligence in mainland China. Reuters had previously reported that Alibaba’s Qwen would be incorporated into the China version of Apple Intelligence, with Baidu technology also in the mix, after the Cyberspace Administration of China registered Apple’s generative AI service in July. Apple briefly posted, then pulled, a Chinese-language Mac guide on connecting Qwen to Siri and Writing Tools.
What the exclusive adds
The new reporting says Apple also trained its own China-specific model with Alibaba’s help, and that this would make Apple the first foreign company approved to offer a proprietary AI model in the country. A suite of AI tools is expected in the coming months. Reuters said it was not immediately clear how Apple’s own model will sit alongside Qwen and Baidu.
- Sources: three people familiar with the matter (unnamed)
- Regulator registration of Apple’s service: July 2026
- Companies: no on-the-record confirmation
Takeaways
- This is sourced reporting, not an Apple announcement
- The dual-track is: Apple’s own model plus local partners
- Feature parity with Apple Intelligence outside China was not specified
Source: Reuters via MacRumors
